机器学习学术速递[8.5]

AI 总结：该研究提出一种LLM自动化流程，可将特征分组为超级节点标注归因图，其效果与人类标注相当，在两跳首都任务中准确率达97%，还可用于开放式探索，推动自动化电路追踪研究。

链接：https://arxiv.org/abs/2608.02632

作者：Ameen Patel, Max Zhang, Nathan Hu

英文摘要：Circuit tracing is an exciting technique for revealing the internal computation of language models, but it requires a time-intensive manual step of grouping individual features or MLP neurons into supernodes. We present a simple pipeline for automating this step: directly presenting feature descriptions to a language model that groups them into supernodes. Using automated interpretability metrics, we confirm that supernodes generated by our pipeline are as interpretable as those generated by human annotators. On a two-hop Capitals task, our pipeline recovers a supernode corresponding to the intermediate hop in 97 of 100 prompts. Finally, we present a simple proof of concept using our pipeline for open-ended exploration, where we automatically annotate 1000 attribution graphs from Wikipedia prompt completions and then use an LLM judge to flag interesting graphs worth human review. We hope this work demonstrates that even simple automation can produce meaningful attribution graph annotations, motivating further work on automated circuit tracing.

AI 总结：该研究用分层持续同调分析，发现预测编码网络（PCNs）的模型规模、简化深度与重构误差、架构类型影响其连通分量合并时机，揭示了PCNs压缩-重构权衡的相关规律。

链接：https://arxiv.org/abs/2608.02816

机构：University of Southern California（南加州大学）

作者：Adam Shaw, Jiayu Li, Michael Sperling, Michael Kim, Alvin Jin

英文摘要：We study the topology of learned representations in predictive coding networks (PCNs), a neuro-inspired bidirectional architecture, using a quantitative layer-wise persistent homology analysis. We train well-performing PCNs on a synthetic classification dataset ($\geq 99.9\%$ test accuracy) and on MNIST ($\geq 95\%$ test accuracy), and measure how topological features change across layers for different architectures and activation functions. We find that smaller PCNs collapse connected components across layers earlier than larger models (Spearman $\unicode{x1D70C} \in [0.72, 0.79]$ across activations), with model size measured as the sum of hidden-layer widths. We also observe a strong negative correlation ($\unicode{x1D70C} = -0.58$) between the depth at which simplification occurs and reconstruction error; i.e., architectures that simplify later reconstruct better. Finally, a seed-level bootstrap comparison across architectures and activations shows that PCNs consistently collapse connected components later than matched MLPs, with an average difference of $3.6$ layers. These results suggest that persistent homology offers a useful quantitative lens on the compression--reconstruction tradeoff in PCNs, and that both model capacity and the recurrent, bidirectional dynamics of predictive coding inference shape when this tradeoff is resolved across layers.

AI 总结：本文提出Maglev，一种带固定记忆的循环Transformer架构，通过耦合模型与记忆一致性损失实现训练可并行性，在验证损失及下游预训练基准上优于基线，参数共享可减少内存。

链接：https://arxiv.org/abs/2608.02870

机构：The University of Texas at Austin（德克萨斯大学奥斯汀分校）

作者：Bo Liu, Qiang Liu

英文摘要：We introduce \ours{}, a recurrent Transformer architecture with fixed-size memory that generalizes sliding-window attention while remaining parallelizable during training. \ours{} consists of two coupled models: a prefiller $Q$, which leverages full attention\footnote{In practice, we use interleaved full and sliding-window attention for $Q$, as this yields stronger performance. The essential requirement is that $Q$ be more expressive than $P$, with access to the full history.} to produce memory targets $m'_t$, and a decoder $P$, which uses only sliding-window attention and recurrent K/V injection to produce decoder memories $m_t$ for next-token prediction. We train \ours{} with a memory consistency loss that aligns $m_t$ with $m'_t$, allowing inference to use $P$ alone. Empirically, \ours{} improves validation loss and downstream pretraining benchmarks over sliding-window and latent recurrent transformer baselines. Moreover, sharing parameters between $P$ and $Q$ reduces parameter memory while preserving most of the gains.

AI 总结： 该研究提出可学习Tsallis图注意力（LTGA），其Tsallis熵指数q可逐边学习，在8个基准测试中验证了其剪枝注意力系数的可解释机制，虽未显著优于调优的α-entmax，但可单次运行替代网格搜索。

链接：https://arxiv.org/abs/2608.02938

机构：University College London（伦敦大学学院）； Holistic AI； University of Utah（犹他大学）

作者：Kleyton da Costa, Bernardo Modenesi

英文摘要：Graph attention normalizes neighborhood scores with softmax, the maximum-entropy choice under Shannon statistics. But homophilic and heterophilic graphs want different attention shapes, and one fixed normalization cannot serve both. We propose \textbf{LTGA} (\textbf{L}earnable \textbf{T}sallis \textbf{G}raph \textbf{A}ttention), a graph attention layer whose Tsallis entropic index $q$ is learned jointly with the weights, interpolating continuously between heavy-tailed ($q\!\!1$) attention at four granularities from a global scalar to a per-edge index, under a bounded reparameterization that starts every model at the GAT baseline. Across eight benchmarks at ten seeds, LTGA-Edge takes the best average rank ($2.75$), but the omnibus test does not reject ($p\!=\!0.199$) and learning $q$ does not beat searching it: a validation-tuned frozen grid reaches $61.4\%$, tuned $\alpha$-entmax $62.2\%$ and a capacity-matched $q\!\equiv\!1$ control $62.0\%$, against $61.7\%$ for LTGA-Edge. What the learned index buys is one run instead of a grid, and an interpretable mechanism: where $q$ leaves $1$, it prunes $42\%$ of attention coefficients to exactly zero, and those edges are selectively the wrong ones, restoring them costs $7.1$ points, while random pruning at the same rate costs $13.0$ more. Project page: this https URL

AI 总结：AcceptMoE 是一种结合目标路由器分数与离线承诺概率的验证器侧专家选择器，可自动调整合格专家数量，在降低流量的同时提升 MoE 投机解码吞吐量，仅轻微损失准确率。

链接：https://arxiv.org/abs/2608.02989

作者：Shuang Liang, Hao (Mark) Chen, Zhiwen Mo, Qianzhou Wang, Guoyu Li, Lingxiao Ma, Wayne Luk

英文摘要：Speculative decoding verifies a tree of draft tokens in one target-model forward pass. For a mixture-of-experts (MoE) target, however, parallel verification can activate the union of the experts selected by all tree nodes, even though only a small subset of those nodes reaches the accepted output. Token count, activated-expert union size, and expert-weight traffic are therefore distinct cost measures: reducing the token workload need not shrink the expert union proportionally, and under offloading, transfer traffic also depends on cache residency. We introduce AcceptMoE, a verifier-side expert selector that combines target-router scores with offline-estimated commitment probabilities and automatically adjusts the number of eligible experts for each verification block, eliminating the need for a user-specified expert budget. Under offloading, AcceptMoE conditions expert eligibility on cache residency instead of predicting natural routes and prefetching the corresponding expert weights. Although constraining target-expert eligibility changes the model distribution, across 12 model-task pairs spanning three MoE targets and four benchmarks, AcceptMoE's mean accuracy is 0.27 percentage points lower than that of EAGLE-3 speculative decoding with natural routing. Served with SGLang at batch size one, it reaches 1.290 times the throughput of this baseline with all expert weights in GPU memory, and 2.06 times under physical expert offloading, while reducing host-to-device traffic by 73.6 percent to 77.1 percent.

AI 总结：本文针对移动端RAG部署的成本问题，提出结合LLM侧查询意图、MoE路由信号与检索分块嵌入的轻量级分块选择方法，提升了排名1的证据选择性能。

链接：https://arxiv.org/abs/2608.03148

作者：Sicong Chang, Yidan Shen, Wen Yu, Jiefu Chen, Xin Fu, Renjie Hu

英文摘要：RAG improves the factual grounding of LLM by incorporating external knowledge, but deploying RAG on mobile and edge devices remains challenging because retrieved context increases computation and memory. A direct way to reduce this cost is to retain only one retrieved chunk before generation, but the top-ranked retrieved chunk is not always the most evidence-supporting one, since retrieval similarity does not necessarily imply evidential sufficiency. Existing context-reduction methods can improve context quality, but often require additional LLMs or compressors that are costly under a strict mobile budget. In this paper, we study lightweight RAG chunk selection as an evidence-alignment problem. Our selector combines three complementary feature sources: question hidden states that represent LLM-side query intent, MoE routing-derived expert signals that capture the generator's internal routing structure, and retrieved chunk embeddings that preserve candidate-side evidence geometry. A compact multilayer perceptron maps these features to an evidence prototype in the chunk embedding space, and the candidate most aligned with this prototype is selected by cosine similarity. For stricter deployment budgets, we further introduce an optional task-aware feature selection strategy to reduce the selector input dimension. To support supervised evaluation, we construct semantic chunk-correctness labels based on evidence sufficiency rather than answer-string containment. Experiments show that the proposed selector consistently improves rank-1 evidence selection over mobile-applicable baselines by an average of 2.5%. These results suggest that using LLM-side query representations and MoE routing information and aligning them with retrieval-side candidate embedding is an effective and parameter-efficient strategy for mobile-applicable RAG chunk selection.

AI 总结：本文提出低秩权重空间消融下的条件坍缩单块理论，通过合成任务实验验证其预测，揭示激活修补与权重空间消融的差异及交互规律。

链接：https://arxiv.org/abs/2608.03620

作者：Abdallah Khemais

英文摘要： Activation patching and weight-space ablation both claim a component is causally responsible for a behavior, yet they act on different objects: one forward pass versus the parameters behind every forward pass. We ask when they agree. We study an idealized model where a conditional computation is carried additively through a residual stream, $F(x)=F_0(x)+\sum_i\alpha_i(x)v_i$, read out by a linear functional, and prove three exact results. First, deleting a subset of carriers collapses a matched input pair onto the same unconditional output \emph{if and only if} the removal is symmetric on the pair and leaves no outside contrast; the error is deterministic, and we give its exact form even when the two conditions hold only approximately. Second, patching a carrier moves the readout by its donor-receiver \emph{contrast}, while ablating it moves the readout by its \emph{absolute level}; neither bounds the other, and we construct pairs where every single-carrier patch flips the decision while no single-carrier ablation does. Third, for an attention head composed with its own layer's normalization and MLP, we derive an exact first-order interaction formula with a provably second-order remainder, vanishing identically when only the MLP is ablated but not, in general, when a head is. Small transformers trained on a synthetic conditional task illustrate all three predictions: across thirty-nine ablation configurations the measured interaction is strongly rank-correlated with the idealized model's predictive accuracy (Spearman $-0.83$), and a second task and architecture reproduces the same pattern, including a further polarity reversal. The single-block interaction result extends past one residual block, and the synthetic validation is tested against a real pretrained model, in a companion paper that takes this theory further along both axes.

AI 总结：UNVaMP是一种结合学生-题目交互与内部记忆的知识追踪方法，纯神经配置UNVaMP-MLP在多数数据集预测性能最优，混合配置UNVaMP-MIRT兼具可解释性且预测成本低，可应用于教育系统并恢复交互潜在结构。

链接：https://arxiv.org/abs/2608.03811

作者：Carson J. Cook, Ahmed J. Zerouali, Anthony Schmidt, Reginald Ziedzor, Paul Lin, Luke G. Eglington

英文摘要：We introduce the Unified Neural Variational Measurement of Proficiency (UNVaMP) architecture, a knowledge tracing method that integrates observed student-item interactions with internal memory to produce evolving latent representations of student knowledge. These representations support accurate predictions of future responses while enabling explicit control over the smoothness of estimated learning trajectories. UNVaMP can be configured as either a purely neural model or a hybrid model that predicts responses through an interpretable measurement function over the latent space. We show that a pure neural configuration (UNVaMP-MLP) achieves the strongest predictive performance among compared models on three out of four datasets. Meanwhile, a hybrid configuration (UNVaMP-MIRT, using a 1PL MIRT measurement function) lags only slightly behind UNVaMP-MLP, indicating that the predictive cost of interpretability is modest. Beyond predictive accuracy, UNVaMP provides the following: a principled mechanism for controlling volatility when estimating student latent variables, quantification of uncertainty over student knowledge state estimates, and flexible input specification that supports heterogeneous student-item interaction features. In addition, the hybrid UNVaMP-MIRT configuration generates interpretable moment-in-time student knowledge state estimates. Using an experimental dataset, we show that auxiliary inputs induce structured changes in the predictive behavior of UNVaMP-MIRT, consistent with sensitivity to underlying structure beyond response correctness. Furthermore, through a simulation study, we show that UNVaMP yields well-behaved knowledge state estimates under controlled measurement conditions. In total, these results indicate that UNVaMP is both useful for real-world education systems and capable of recovering underlying structure from student-item interactions.

AI 总结：针对大语言模型家族切换时需重新预填充的问题，提出跨模型KV缓存迁移方法，通过闭式线性映射器复用源模型KV缓存，可提升预填充速度，保留大部分准确率，具备实用性。

链接：https://arxiv.org/abs/2608.03893

作者：Taekyung Heo, Rasoul Shafipour, Ritchie Zhao, Maximilian Golub, Mohammad Mahdi Kamani, Ritika Borkar, Makesh Tarun Chandran, Pantea Zardoshti, Bita Darvish Rouhani

英文摘要：Production deployments often swap between different-sized models in a family for cost-quality cascading, mid-conversation switching, and routing, and each swap forces the receiver to repay the prefill from scratch. We propose cross-model KV cache transfer, where the receiver reuses the source's KV cache, skipping prefill. We find that cross-model KV has substantial linear structure across matched-KV pairs, where source and target share KV head count and per-head dimension. On Qwen3 14B->32B, one source layer explains 56% of variance in the target's keys and 32% in values, rising to 79% and 65% with multiple source layers. Building on this, we design a closed-form ridge mapper that operates per head and proceeds in three steps. First, for each target layer we select the top-k most predictive source layers and concatenate their KV as input. Second, we strip RoPE from the keys before mapping, so the fit is position-free and reusable across context lengths. Third, we fit ridge regression on a small calibration set of 500 FineWeb-Edu sequences of 1,024 tokens each. Surprisingly, across six pairs in three families, this linear mapper retains 73-98% of the receiver's standalone-prefill accuracy on four pairs, while two degrade sharply. A nonlinear MLP recovers up to +37 pp HellaSwag retention on the failures. The mapper runs 2.7-25x faster than re-prefill and remains stable across multi-turn handoff, making cross-model KV cache transfer practical.

AI 总结：提出Sparse Weight Decomposition（SWD）方法，将预训练Transformer权重矩阵分解为稀疏因子以提供可解释电路单元，在低数据量下实现高保真电路提取，适用于多模型及全模型替换，还支持零数据变体以拓展机械可解释性分析应用。

链接：https://arxiv.org/abs/2608.03913

作者：Chuanhao Yan, Xuhan Huang, Yawen Duan, Zhenfei Yin, Hang Zhao, Bryan Dai, Jie Fu

英文摘要： Dense pretrained transformers do not naturally expose interpretable units for circuit extraction. Existing approaches obtain such units by learning auxiliary sparse representations or training sparse models, incurring substantial additional computation while potentially introducing a fidelity gap between the representation being analyzed and the original pretrained model. We propose Sparse Weight Decomposition (SWD), which reparameterizes pretrained linear projections by factorizing each weight matrix into two sparse factors whose shared intermediate coordinates serve as individually addressable circuit units. Without training a separate replacement network, this parametric representation supports the same scoring, selection, and ablation circuit extraction workflow used for methods that learn sparse features. Across single-matrix replacements, SWD matches the held-out fidelity achieved by Transcoder and other strong baselines while using less than 1% of the data that those baselines use to train their replacements. For matched replacement fidelity, SWD reaches the same circuit sufficiency and necessity targets with fewer active read/write edges and selected units across tasks on GPT-2, Qwen2.5, and Qwen3.5-27B. We further show that SWD remains effective for full-model replacement of all attention and MLP weight matrices after fine-tuning the nonzero factor values. Finally, SWD also features a zero-data variant, allowing broader use of mechanistic interpretability analysis (e.g., per-step analysis).

AI 总结：本文对比缪昂与AdamW在Mamba-2 1.3亿参数模型上的表现，发现仅在输出投影使用缪昂可提升令牌效率，该优势与条件数无关。

链接：https://arxiv.org/abs/2608.03941

作者：Arslan Battalov, Karim Kramin, Alexander Markotenko, Sofia Sinitsina

英文摘要：Muon is a recent optimizer that orthogonalizes the update to each weight matrix with a Newton-Schulz iteration, which performs steepest descent under the spectral norm. Almost all the evidence for it comes from Transformer models, and its behavior on state-space models is largely unreported. We compare Muon with AdamW on Mamba-2 130M under a controlled protocol that varies only which weight groups are trained with Muon. The benefit is localized. Muon on the output projection alone beats Muon on the input projection or on both. The advantage is mainly one of token efficiency. It holds on two corpora and two token budgets, and persists when training continues well past the compute-optimal point. Conditioning does not explain the gain. Muon lowers the condition number of whichever projection it trains, but the better-conditioned input projection is not the one that helps.

AI 总结：本文提出结构保留的PhenMol框架，通过解耦分子与细胞表示实现表型引导对齐，在多类任务上提升性能，为药物发现整合细胞表型与化学知识提供了有效途径。

链接：https://arxiv.org/abs/2608.02688

机构：Xiangtan University（湘潭大学）； Hunan University（湖南大学）； Beijing University of Posts and Telecommunications（北京邮电大学）； Southeast University（东南大学）

作者：Xuan Lin, Jingyu Sheng, Tengfei Ma, Li Sun, Dapeng Xiong

英文摘要：Phenotypic drug discovery enables the discovery of functional relationships between molecular structures and cellular responses. However, existing multimodal representation learning methods often optimize cross-modal alignment without considering the intrinsic organization of chemical space, resulting in distorted molecular representations and loss of structural information. We propose \textbf{PhenMol}, a structure-preserving framework for phenotype-aware molecular representation learning. PhenMol disentangles molecular and cellular representations into shared and private components, enabling phenotype-guided alignment while preserving chemical structures through a dedicated molecular branch. This design integrates cellular phenotype information without disrupting molecular neighborhood organization. Experiments on approximately $3.04 \times 10^{4}$ molecule--cell morphology pairs demonstrate that PhenMol improves molecular property prediction across 270 bioactivity tasks, molecule--phenotype retrieval, and clinical trial outcome prediction. Moreover, ECFP4-based structural analysis shows that PhenMol better preserves molecular neighborhoods and reduces embedding distortion compared with existing multimodal alignment methods. These results highlight the importance of structure-aware constraints in multimodal molecular representation learning and provide an effective approach for integrating cellular phenotypes with chemical knowledge for drug discovery.

AI 总结：该研究针对柔性作业车间调度中深度强化学习模型参数多、推理慢的问题，提出PLAN框架，通过解耦状态演化与上下文聚合实现高效学习，在三类基准任务上优于基线，参数量仅为基线的22%-47%且延迟显著降低。

链接：https://arxiv.org/abs/2608.03041

作者：Dhivya Dharshini Kannan, Wei Zhang, Jieyi Bi, Yingpeng Du, Tianjun Wei, Jie Zhang, Zuming Liu, Anupam Trivedi

英文摘要：Deep reinforcement learning (DRL) approaches for flexible job shop scheduling (FJSP) heavily rely on attention-centric architectures to achieve state-of-the-art performance. However, these models suffer from excessive parameter counts and prohibitive inference latency as problem scales expand. While liquid neural networks (LNNs) offer a parameter-efficient alternative for modeling adaptive state evolution, their inherently sequential dynamics bottleneck computational efficiency. To resolve this trade-off, we propose PLAN (Parallel Liquid-inspired Approximation Network), a lightweight representation learning framework that reformulates continuous liquid-state dynamics into a discretized and parallelizable formulation. PLAN structurally decouples state evolution from context aggregation, where liquid-inspired updates handle the primary evolving state representation, and a lightweight context aggregation module provides complementary global context. Furthermore, PLAN acts as a versatile, plug-and-play backbone that generalizes to complex FJSP variants, pairing with a compact stochastic module for stochastic FJSP and replacing heavy heterogeneous graph transformers in multi-faceted dynamic FJSP. Extensive evaluations across deterministic, stochastic, and multi-faceted dynamic FJSP benchmarks show that PLAN reduces the average makespan by 1.2%, 1.4%, and 2.3%, respectively, compared with the corresponding state-of-the-art baselines, with the improvement reaching 10.2% in one benchmark setting. PLAN also reduces average inference latency by 13.2%, 31.7%, and 26.9%, respectively, with a maximum reduction of 69.2% on the largest instances, while using only 22$-$47% of the baseline parameters.

AI 总结：该研究提出无需训练的KV缓存索引SAKI，通过优化注意力分数目标，在多个开源大模型上的长上下文KV检索任务中，相比键PCA方法显著提升了前k个召回率。

链接：https://arxiv.org/abs/2608.03228

作者：Lin Zhang

英文摘要： Existing low rank KV cache methods preserve either model weights or key variance, neither of which directly reflects the attention scores used during inference. We derive the expected attention score distortion caused by rank r key compression and show that it yields a covariance weighted low rank objective. Under a margin condition, controlling this distortion also improves top k recall. The optimal rank r solution has a closed form asymmetric factorization obtained from the SVD of the covariance weighted query key operator. This motivates SAKI, a training free KV cache index that directly preserves attention scores rather than key reconstruction quality. Across LLaMA 3.1 8B, Qwen 2.5 7B, Mistral 7B v0.1, and Llama 3.2 3B, SAKI outperforms key PCA at every tested rank. At rank 32, it removes 13 to 30 percent of PCA's remaining top 64 recall error, including improvements from 0.748 to 0.799 on LLaMA 3.1 8B and from 0.786 to 0.850 on Qwen 2.5 7B. It improves 68 to 89 percent of attention heads per model, with the largest gains in deeper layers. Predicted score MSE reductions closely match empirical measurements, with a Pearson correlation of 0.997, while ablation studies confirm that the gains arise from optimizing the attention score objective rather than covariance weighting alone. Analysis of the scoring operator further explains why weight only, invariant subspace, and key reconstruction methods can be suboptimal.

AI 总结：该研究提出双语义化学嵌入模型CheMatE，通过两阶段训练将SMILES与自然语言联合表示，在分子性质预测等下游任务上表现具竞争力。

链接：https://arxiv.org/abs/2608.03855

作者：David Ming Segura, Jeremy Goumaz, Joshua W. Sin, Bojana Ranković, Philippe Schwaller

英文摘要：Transformer models have revolutionized natural language processing (NLP), and text-based molecular representations like SMILES have successfully extended these architectures to chemistry. However, domain-adaptive pre-training often causes models to overfit to chemical syntax, catastrophically forgetting their foundational semantic capabilities. To address this challenge, we introduce CheMatE, a chemistry-oriented embedding model that jointly captures molecular structure and domain-specific natural language within the same representation space. Built on a ModernBERT backbone, CheMatE learns bi-semantic representations through a two-stage training procedure: continued masked language modeling (MLM) followed by a Matryoshka contrastive learning stage via Multiple Negative Ranking Loss (MNRL). First, we train the model using MLM on a novel, large-scale corpus of SMILES-annotated, long-context scientific documents that were constructed and curated from FineWeb and ChemPile (comprising 10.4B and 11.5B tokens, respectively). Subsequently, the model undergoes contrastive learning using a synthetic dataset of SMILES-text pairs algorithmically derived from our original training corpus. This design exposes the model to SMILES-enriched scientific literature, enabling bi-semantic understanding. We evaluate CheMatE across a range of downstream tasks covering molecular property prediction and scientific language understanding. Our results demonstrate that coupling our custom-curated datasets with this sequential training strategy yields robust, highly transferable representations. By effectively unifying structural and contextual signals within a single text-based framework, CheMatE achieves competitive performance across both specialized chemistry models and general-purpose language model baselines.

AI 总结：该研究将在线策略自蒸馏扩展至开放式生成，提出以评分规则作为特权信息（RuPI），在Qwen、Llama等模型上，其性能优于评分规则作为奖励的RL及参考完成PI蒸馏方法。

链接：https://arxiv.org/abs/2608.02948

机构：Apple（苹果公司）

作者：Deepika Bablani, Ajay Gupta, Wanming Chen

英文摘要：On-policy self-distillation (OPSD), where a single model acts as both student and teacher with different contexts, has shown promise in verifiable domains like math, where hard privileged information (PI) in the form of ground-truth answers structurally constrains valid continuations. We extend OPSD to open-ended generation using soft PI in the form of rubrics that guide preferences but admit many valid responses. Rubrics have served as scalar rewards for reinforcement learning (RL); we show that they provide substantially richer signal as dense PI for distillation, and contrary to intuition, soft rubric PI provides a larger and more effective training signal on student roll-outs than hard reference completion PI in this regime. A reference completion is one point in a set of valid responses, so distilling towards it over-constrains the student, while rubrics specify the preference structure shared across the set of valid responses. We show the effectiveness of using rubrics as PI for open-ended generation across Qwen and Llama model families and show that it outperforms rubric-as-reward (RaR) RL using HealthBench, a benchmark that grades open-ended health responses against physician-created rubrics, providing dense token-level supervision for open-ended tasks; RuPI beats RaR RL by up to +0.10 absolute score and, under matched recipe and KL direction, beats reference-PI by +0.034 to +0.079 absolute score across three models. We further show that these findings generalize to training on the RubricHub Science corpus and evaluating on ResearchQA: soft rubric PI outperforms both reference-PI distillation and RaR RL (66.6% vs. 64.2% and 57.6%).

AI 总结：SP3O是一种无需奖励建模的新型PbRL算法，利用分段级偏好反馈，通过PPO型损失函数优化策略，在机器人控制和LLM微调等长视界任务中性能优于现有算法。

链接：https://arxiv.org/abs/2608.02951

机构：University of Michigan（密歇根大学）

作者：Evan Assmus, Qining Zhang, Lei Ying

英文摘要： Preference-based reinforcement learning (PbRL) for general stochastic MDPs often requires training a reward model. Existing reward-model-free methods are either restricted to bandits or deterministic MDPs, such as DPO or P3O, or use zeroth-order, gradient-free optimization, which in general exhibits a slower convergence rate than gradient-based algorithms. Furthermore, existing reward-model-free preference-based RL algorithms almost exclusively use trajectory-level feedback, which can require significant effort from a human evaluator when trajectories are long. On the other hand, segments are much shorter, so they are easier to compare and evaluate. In this paper, we introduce a novel reward-model-free, critic-free, and gradient-based PbRL algorithm compatible with segment preferences named Segment Pairwise Proximal Policy Optimization (SP3O). SP3O utilizes segment-level preference feedback to construct an accurate policy value difference estimator via off-policy importance sampling, and then uses the estimator to compute the policy gradient via a PPO-type loss function. We provide a theoretical basis for the algorithm and analyze the tradeoff in choosing the segment length. We also evaluate it experimentally against other PbRL/RLHF algorithms in robotic control and LLM finetuning settings to show its improved performance, especially in long-horizon tasks.

AI 总结：本研究针对Q学习中TD目标聚合受估计噪声放大误差的问题，提出SADQ方法，通过动力学模型的单步展开预测正则化TD目标，在多类任务上提升了DQN的训练稳定性。

链接：https://arxiv.org/abs/2608.03069

作者：Lipeng Zu, Xiaonan Zhang

英文摘要：Deep Q-Networks (DQNs) learn value functions through bootstrapped temporal-difference updates, where future returns are approximated using a greedy maximization over next-state action values. While effective, this aggregation rule is inherently sensitive to estimation noise: when Q-values are uncertain, the maximization operator deterministically favors the largest estimate, regardless of its reliability, leading to amplified errors through bootstrapping. In this work, we propose the \textbf{S}uccessor Rollout \textbf{A}ggregation \textbf{D}eep \textbf{Q}-Network (SADQ), a simple modification to Q-learning that regularizes how the TD target is formed. SADQ uses one-step rollout predictions from a learned dynamics model to guide the comparison among candidate next-state actions, introducing additional structure into the aggregation step without altering the underlying learning framework. The resulting mixed Bellman update attenuates unreliable maxima while preserving the standard fixed point under diminishing model error. We provide theoretical analysis showing that SADQ reduces bootstrap-induced overestimation in a pointwise manner. Empirically, SADQ consistently improves training stability across classical control tasks, real-world vector-based environments, and Atari benchmarks when compared to strong DQN variants.

AI 总结：针对多奖励强化学习中GDPO难以平衡不同粒度奖励信号的问题，提出SMOPD两阶段训练方法，在1.5B、3B、7B骨干模型上的互补与冲突奖励设置下性能优于GDPO。

链接：https://arxiv.org/abs/2608.03092

作者：Wen Wang, Jiahua Bao, Tu Yongsiqi, Yihao Liu, Haotian Zhou, Haoxuan Ma, Mengyu Zhou, Wenkui Fan, Junwei He, Xiaoxi Jiang, Guanjun Jiang

英文摘要：We aim to improve model performance in multi-reward reinforcement learning training process. Existing Group reward-Decoupled Normalization Policy Optimization (GDPO) has mitigated the issue of reward signals masking one another during direct scalarization by normalizing each reward dimension separately before aggregation. However, our experiments show that GDPO still struggles to balance reward signals with different granularities. Specifically, in some particular training tasks, the model may receive a dense reward that assigns fine-grained scores ranging from 0.1 to 1.0, together with a sparse reward that provides only binary feedback of either 0 or 1. In such cases, we find that the sparse reward may provide an insufficient optimization signal, preventing its corresponding capability from being effectively reinforced. Therefore, how can we strengthen the optimization signal from the sparse reward without sacrificing the capability already learned from the fine-grained reward? To overcome this limitation, we propose Specialize-and-Merge Online Policy Distillation (SMOPD), a two-stage training method for multi-reward optimization. Stage1-Specialize: SMOPD first employs reward-priority configurations to train multiple reward-specialized teachers, allowing each reward to be learned under conditions where its signal can effectively drive optimization. Stage2-Merge: SMOPD then utilizes online policy distillation to combine the reward-specialized capabilities of these teachers into a single student policy, while maintaining balanced task-level optimization. To validate our method, we conduct experiments on two multi-reward settings: complementary rewards(tool-calling accuracy and format) and conflicting rewards (helpful and harmless rewards). Based on above settings, SMOPD outperforms GDPO across 1.5B, 3B and 7B backbones.

AI 总结：针对离线强化学习中分布外动作估计误差被放大的问题，提出CSDG算法，通过分离样本内参考与局部修正项实现泛化控制，经实验在Gym-MuJoCo等任务上表现强劲。

链接：https://arxiv.org/abs/2608.03108

作者：Yi Yang, Zhennan Chen, Mingfeng Lv, Hanlei Li, Zhengsen Ruan, Lvqing Yang

英文摘要：Offline reinforcement learning (offline RL) can benefit from nearby out-of-distribution (OOD) actions, but estimation errors at these actions may be amplified by bootstrapping. Existing regularization and local-generalization methods control either the admissible OOD region or the influence of generalized targets, often through separate mechanisms. We propose Convex Hull Neighborhood Smooth Dual Generalization (CSDG), which expresses the Bellman backup as an in-sample value target plus a CHN-local correction. This formulation makes the generalized contribution explicit and separates it from the in-sample reference path. The correction is obtained by smoothing in-sample-oriented and OOD-oriented candidates sampled at different perturbation radii. A mixture coefficient lambda scales its contribution to each backup, while the recursive discount remains gamma. Under boundedness and fixed perturbation kernels, we derive an exact one-step correction identity, a time-varying iterate bound, and a fixed-point bound that depends only on the branch discrepancy at the fixed point. We further characterize the implicit policies induced by the idealized operators and give a conditional non-degradation criterion. The practical algorithm approximates these quantities using asymmetric bounded noise and expectile regression, without exact support classification or an additional pessimistic OOD penalty. Experiments on Gym-MuJoCo and AntMaze show strong aggregate performance and stable value estimation. Code is available at: this https URL

AI 总结：针对智能体强化学习中稀疏奖励无法分配中间决策信用的问题，提出ADRS框架，通过校准教师分数等提升多轮语言智能体在长时域任务的性能，且增益稳定。

链接：https://arxiv.org/abs/2608.03223

作者： Ranxu Zhang, Guinan Chen, Chenshaodong, Jinghao Lin, Xiaozhou Xu, Sunzhe, Yanyong Zhang, Chao Wang

英文摘要：Agentic reinforcement learning enables LLM agents to learn through interaction, but sparse trajectory-level rewards reveal success without identifying which intermediate decisions deserve credit. Training-only privileged skills can provide denser supervision by allowing the same frozen policy snapshot to rescore fixed tokens from skill-free trajectories while conditioned on task-matched procedural skills. Existing methods, however, do not jointly calibrate teacher scores across interaction steps, relate teacher confidence to realized returns, and integrate the resulting signal into native reward-to-advantage construction. We introduce Agentic Reinforcement Learning with Self-Distilled Reward Shaping (ADRS), a framework for constructing return-associated token-level credit for multi-turn language agents. ADRS centers and normalizes privileged token scores within each step, modulates them with a return-associated Teacher Value Advantage (TVA) gate based on within-group confidence--return association, and incorporates the gated token signal into native RL credit construction. Together, these components determine what the teacher prefers, when that preference is return-relevant, and how it enters the native reinforcement-learning credit path, while keeping rollouts and inference skill-free. Finally, experiments across three interactive benchmarks show that ADRS consistently improves performance on long-horizon tasks, with gains persisting across RL backbones, reduced-data settings, unseen tasks, and extended training. For anonymous review, our code is available at the following the link: this https URL

AI 总结：提出TimeRLM（基于递归语言模型的时序异常定位框架），结合强化学习后训练后，在合成基准AnomalyXL及真实世界数据上，均优于现有时序语言模型，可实现长时序数据的精准异常定位。

链接：https://arxiv.org/abs/2608.03391

作者：Nicolas Zumarraga, Lorenzo Steno, Ning Wang, Max Rosenblattl, Thomas Kaar, Maxwell A. Xu, Kevin O'Sullivan, Markus Kreft, Elgar Fleisch, Paul Schmiedmayer, Patrick Langer, Robert Jakob

英文摘要：Precise anomaly localization over long-context time series is a crucial task in monitoring applications across clinical care, industrial operations, financial services, and logistics, where brief evidence may hide inside long spans of high-frequency data. Time-Series Language Models (TSLMs) are able to ingest time series data and verbalize findings on anomalies in natural language; however, recent benchmarks report a decrease in retrieval performance at long contexts, mirroring failure modes in text, vision, and audio. In the text domain, Recursive Language Models (RLMs) can recover much of this lost performance by keeping context external to the large language model (LLM), allowing the model to query it through code. We present TimeRLM, an RLM formulation for time-series that sequentially manipulates the signal using code and vision capabilities. We further introduce AnomalyXL, a synthetic long-context anomaly localization benchmark with programmatically injected anomalies that require precise retrieval. We implement five different task categories and two variants: AnomalyXL-MCQ and AnomalyXL-Localize. TimeRLM outperforms every evaluated TSLM and single-pass baseline on four of the five AnomalyXL-Localize tasks, reaching 0.682 IoU on localization and 0.745 on classify-with-evidence, versus at most 0.329 and 0.072 across all baselines. We post-train TimeRLM using reinforcement learning. The resulting model further improves performance and requires approximately one-third as many agent interaction turns as its untrained base model to produce a final answer. On unseen real-world ECG, sleep and software observability recordings, the post-trained TimeRLM retains or improves performance, surpassing TSLMs despite being trained exclusively on synthetic data. Our findings suggest recursive interaction with time-series is an effective approach for long-horizon retrieval.

AI 总结：针对通用效用强化学习的部署效用误指定问题，提出极小极大学习框架及两种收敛随机算法，经LLM安全对齐等实验验证了其有效性。

链接：https://arxiv.org/abs/2608.03562

作者：Zixuan Liu, Fangzheng Wu, Brian Summa, Zizhan Zheng

英文摘要：Reinforcement learning (RL) with general utility extends classic RL by optimizing an arbitrary utility functional of the policy-induced occupancy measure, thereby enabling a broader range of applications. However, previous work on general utility RL typically assumes the evaluation utility is fixed and correctly specified. In practice, the utility used at deployment can deviate from the training one, creating a robustness gap that prior work does not address. Motivated by this, we propose robust general-utility RL, a minimax learning framework that trains policies against utility misspecification within a prescribed uncertainty set. Our framework strictly generalizes standard general-utility RL while also providing a unified view of many existing RL frameworks, including reward-robust RL and constrained RL, through appropriate choices of the utility uncertainty set. We further develop provably convergent stochastic algorithms for two regimes. For concave utilities, we develop a projected stochastic gradient descent-ascent method and establish stationarity guarantees. For the more challenging nonconcave regime, we propose a stochastic prox-extragradient algorithm that mitigates ill-posed behavior induced by nonconcavity, with convergence guarantees to approximate first-order stationarity. Experiments on LLM safety alignment and exploration maximization tasks further corroborate the convergence behavior consistent with our theory.

AI 总结：CausalOPD 是一种课程在线过程提炼框架，通过识别首错误步并结合短视界强化学习，提升了学生模型的因果链推理能力，在三个领域中表现优于序列级提炼及专有参考模型。

链接：https://arxiv.org/abs/2608.03673

作者：Jian Zhang, Bingyi Wang, Yizhi Liu

英文摘要： Many critical reasoning tasks, including clinical diagnosis, legal judgment, and industrial fault diagnosis, require step-dependent causal chains in which early errors propagate and correct conclusions can mask invalid reasoning. Although large language models perform well on such tasks, privacy, latency, and controllability motivate distillation into locally deployable models. Standard trajectory imitation does not correct process errors on the student's own rollout distribution. We propose CausalOPD, a curriculum online process distillation framework. A knowledge-augmented teacher first provides trajectories grounded in domain-specific causal rules, entity relations, and structural constraints. The student then generates on-policy trajectories, and the teacher identifies the first wrong step, defined as the earliest transition that verifiably violates available constraints. Starting from the verified prefix, short-horizon reinforcement learning repairs this localized failure. A causal-stage curriculum advances from evidence-level to mechanism-level and conclusion-level errors, following their propagation order. Across three domains, CausalOPD improves average path correctness by 23.4 percentage points over sequence-level online process distillation and reduces the right-label-wrong-reasoning rate from 15.7% to 4.4%. The domain-specific 8B students also surpass both evaluated proprietary references in path correctness across all domains.

AI 总结：本文提出DiagLoop反事实数据飞轮，利用合成场景训练8B诊断大语言模型，在工业系统与疾病类别上的严格路径正确性及对照性能均优于传统基线与专有参考模型。

链接：https://arxiv.org/abs/2608.03674

作者：Jian Zhang, Bingyi Wang, Yizhi Liu

英文摘要：Causal diagnostic models must explain how conclusions follow from evidence because diagnoses guide repairs and treatments. Yet serious cases are scarce, records rarely contain reasoning paths, and data transfer poorly across configurations, complicating local deployment. We present DiagLoop, a counterfactual data flywheel that converts codified physical relations or clinical guidelines, authored once per mechanism family, into training supervision beyond recorded cases. A training-only teacher proposes counterfactual worlds by varying causes, contexts, and observations, while an independent hybrid checker admits only valid worlds. The student reasons through symptom abstraction, causal-chain construction, and root-cause attribution. Stage-specific criteria identify its earliest failure. For nonterminal failures, a bounded repair probes downstream competence, and the resulting weakness profile guides subsequent data generation. Stage-localized reinforcement learning updates only the model-generated continuation, while replay and preservation reduce forgetting. The same criteria govern admission, attribution, reward, and regeneration through checks separate from the proposer. Using only synthesized scenarios and no case-level expert reasoning annotations, the resulting 8B model improves strict path correctness over the strongest conventional baseline. Gains are 11.6 points across eight industrial systems and 5.5 points across ten disease categories. Gains over a deranged-routing control are 3.9 and 2.3 points, respectively. The model also exceeds the evaluated proprietary references in both domains, even when they receive few-shot examples or the specification in context.

AI 总结：FedCritic-MIMO是面向开放解耦6G RAN的通信高效无服务器联邦多智能体强化学习框架，通过稀疏评论者交换等技术实现多小区大规模MIMO资源控制，性能最优且通信开销降低76%。

链接：https://arxiv.org/abs/2608.03852

作者：Amin Farajzadeh, Melike Erol-Kantarci

英文摘要：This paper proposes FedCritic-MIMO, a communication-efficient serverless federated multi-agent reinforcement learning framework for AI-native resource control across independently deployable cell-level controllers in open and disaggregated 6G RANs. Controllers share no trainer, retain local actors and personalized critic components, and exchange only compatible shared critic parameters. FedCritic-MIMO targets reuse-$1$ multi-cell massive-MIMO OFDMA deployments, where RAN controllers jointly manage user scheduling, per-stream power allocation, beamforming, interference, and long-term QoS with limited inter-controller signaling. Each base station locally executes its actor without centralized training or actor federation, while critic knowledge is exchanged peer-to-peer over an interference-aware graph. It enables this collaboration through wireless-aware event triggering, adaptive layer-wise top-$k$ sparse critic exchange with error feedback, and balanced interference-aware fusion. We establish conditional finite-time stationarity and consensus guarantees for the balanced, compressed peer-to-peer critic recursion under a fixed-policy, frozen-target critic-regression model. In strongly interference-coupled reuse-$1$ simulations, FedCritic-MIMO achieves the best performance-communication tradeoff among heuristic, independent-learning, centralized-training, and communication-ablation baselines. It achieves the highest held-out throughput, improves user-rate distribution and mean SINR, increases QoS satisfaction, and attains the lowest interference cost per delivered bit among learning baselines. It reduces critic-communication overhead by $76\%$ relative to uncompressed distributed critic exchange. These results demonstrate that serverless exchange of compatible shared critic parameters can coordinate RAN controllers without centralized trajectory collection or parameter-server aggregation.

AI 总结：本研究提出结构感知微调（SAFT）方法，利用LoRA适配器和内在结构先验优化视觉语言模型奖励信号，实现更快策略收敛与更好对齐，为稳定文本条件强化学习提供可扩展路径。

链接：https://arxiv.org/abs/2608.03875

作者：Pyrros Koussios, Chenhao Li, Xin Chen, Andreas Krause

英文摘要：Designing effective reward functions remains a major bottleneck in Reinforcement Learning (RL). Recent work uses large foundation Vision-Language Models (VLMs) as reward models, computing text-observation similarity to bypass manual reward engineering. Although promising, these rewards are often noisy and unreliable, limiting their direct utility during deployment. We present Structure-Aware Fine-Tuning (SAFT), a simple, self-supervised method that refines these imperfect reward signals online without access to ground-truth supervision. SAFT leverages intrinsic structural priors to regularize the VLM's latent space via LoRA adapters. We rigorously evaluate SAFT across a spectrum of base model capabilities to demonstrate its versatility. Our results show that SAFT consistently denoises the reward landscape, yielding faster policy convergence and substantially improved alignment (EPIC distance) relative to the underlying base model, suggesting that failures can often be attributed to structural brittleness rather than semantic misunderstanding. By replacing extensive human preference annotation with structural inductive biases inherent to the task, SAFT offers a scalable path for stabilizing text-conditioned RL and underscores the broader value of incorporating task structure as a general inductive bias.

AI 总结：针对扩散模型偏好对齐的时间信用分配挑战，提出Latent Reward Registers机制，结合RG-OPD和RGS策略，在高噪声下实现最优性能且大幅降低计算成本。

链接：https://arxiv.org/abs/2608.03929

作者：Yuanshen Guan, Zipeng Feng, Zhiwei Xiong, Peiqin Sun

英文摘要：Aligning diffusion models with human preferences usually relies on a sparse terminal reward evaluated on the final generated samples, presenting a severe temporal credit-assignment challenge across the multi-step denoising process. We propose Latent Reward Registers, a mechanism that estimates terminal preference directly from intermediate noisy latents by prepending learnable, position-free register tokens to the input sequence of a frozen Diffusion Transformer (DiT). This independent readout mechanism extracts latent reward evidence without altering the generator's hidden states or velocity field. The resulting dense, differentiable reward signal throughout the full denoising process facilitates two alignment strategies. For training, Reward-Gradient On-Policy Distillation (RG-OPD) distills reward-guided updates along on-policy trajectories, bypassing the computationally expensive rollouts of standard policy gradients. For inference, Reward-Guided Sampling (RGS) steers trajectories via magnitude-matched reward gradients without parameter updates. Empirically, at high noise levels (u = 0.8), the registers reach the highest pairwise accuracy among the evaluated latent reward models. Furthermore, RG-OPD outperforms online reinforcement learning baselines while reducing GPU hours by up to 33x, and RGS establishes a new state-of-the-art among training-free methods, strictly enhancing both alignment and perceptual metrics. Code and weights are available at this https URL

AI 总结：本文提出GoT-CD方法，其生成的DAG在因果发现基准中表现优异，但研究发现事后路径特定公平性审计对因果发现的结构误差较为脆弱，需结合结构发现开展路径特定公平性分析。

链接：https://arxiv.org/abs/2608.02877

机构：University of California, Irvine（加利福尼亚大学欧文分校）； University of Orléans（奥尔良大学）

作者：Nitish Nagesh, Elahe Khatibi, Thomas Dean Hughes, Mahdi Bagheri, Pratik Gajane, Amir M. Rahmani

英文摘要：Causal discovery recovers directed structure from observational data and is increasingly used in clinical settings to support mechanism reasoning and fairness audits of predictive models. Path-specific counterfactual fairness asks whether a protected attribute influences an outcome through illegitimate pathways, but these estimands are defined relative to a supplied causal graph and therefore inherit whatever errors the discovery step introduces. Discovery methods are routinely scored on aggregate structural metrics that weight all edges equally, and no established evaluation asks whether the specific pathway an audit depends on survives discovery---or what the audit reports when that pathway is missing. Here we show that full-graph Graph-of-Thoughts reasoning yields acyclic discovered graphs that are structurally competitive with large language model (LLM) baselines, yet that structural fidelity alone does not guarantee fairness-faithful audits. We introduce GoT-CD, in which the reasoning unit is a complete candidate edge set: multiple graphs are generated in parallel, scored by a deterministic validity function, and merged under a hard union constraint that forbids invented edges, with greedy projection enforcing a DAG before commitment. GoT-CD returns a valid DAG on all five reported benchmarks and achieves the best DAG-valid F1 score among LLM methods on Asia, Alzheimer's, and COVID-Respiratory datasets. On an Alzheimer's benchmark with known unfair path, a post-hoc path-specific audit shows that five of eight discovered graphs recover no path from the sensitive attribute to the outcome and therefore report a null overall effect while mediated effects persist, necessitating downstream path-specific fairness analysis along with structural discovery.

AI 总结：该研究针对序列决策反事实推理的确定性因果模型假设与MDP随机性的矛盾，提出概率非确定性因果模型下的鲁棒反事实策略优化方法，经脓毒症治疗模拟器验证有效。

链接：https://arxiv.org/abs/2608.02893

机构：Durham University（杜伦大学）； University College London（伦敦大学学院）

作者：Jessica Lally, Milad Kazemi, Nicola Paoletti, David Watson, Sander Beckers

英文摘要：Counterfactual inference approaches for sequential decision-making typically assume deterministic causal models, where all randomness stems from latent variables. However, Markov Decision Processes (MDPs) are inherently stochastic. We address this by formalising counterfactual policy optimisation under probabilistic nondeterministic causal models, which properly separates latent confounding from irreducible stochasticity, and here propose a first practical optimisation problem for identifying robust counterfactual policies under a sensitivity analysis framework. We validate our approach on a sepsis treatment simulator, where diabetes status acts as a hidden global confounder.

AI 总结：本研究针对单细胞基因表达向量的自监督生成任务，扩展因果Transformer模型，发现单细胞基础模型的双指数缩放定律与计算最优前沿，还探讨其微调用于扰动响应预测的可能。

链接：https://arxiv.org/abs/2608.02961

作者：Aleksandr Sharipov, Yusif Mukhtarov, Igor Molybog

英文摘要：We study a self-supervised generation task for single-cell gene expression vectors: given a set of vectors from a cell type, we aim to generate additional gene expression vectors of that cell type. For this task we characterize both the biological fidelity of the generated gene expression vectors and the scaling behavior of the pretraining loss. The model is a causal transformer paired with a learned quantized VAE tokenizer, trained with a cross-entropy loss. To evaluate the model, we condition it on held-out gene expression vectors of a cell type and generate vectors of gene expression, comparing the resulting distribution over gene expression vectors to the ground truth distribution of that cell type. We study the scaling properties of the proposed architecture by varying the number of trained parameters and the amount of training data. To our knowledge, we find the first jointly-fit two-exponent scaling law and compute-optimal frontier for a single-cell foundation model. Finally, we discuss how this pretrained model could be finetuned for perturbation response prediction.

AI 总结：本文提出一种参考过程设计框架，可同时实现无模拟训练与有限时间生成，还揭示得分匹配并非扩散模型训练的基础，且条件流匹配是该框架的小噪声极限。

链接：https://arxiv.org/abs/2608.03117

作者：Kentaro Kaba, Masayuki Ohzeki, Yuki Sughiyama

英文摘要：The performance of generative diffusion models is determined by the choice of the reference diffusion process connecting the empirical and prior distributions. Conventional approaches typically trade off simulation-free training against finite-time generation. We propose a framework for designing the reference process that achieves both simultaneously. The key idea is to prescribe tractable time-dependent conditional distributions and then construct the reference process realizing them as its marginals. This framework reveals that score matching is not fundamental to diffusion-model training but instead emerges naturally through reversal of the reference process. We further show that conditional flow matching arises as the small-noise limit of the proposed framework.

AI 总结：Any-OPD是首个针对任意潜在流匹配生成器对的策略上蒸馏框架，通过表示空间桥接实现异构模型间的蒸馏，以五分之一规模达到教师模型性能，解决了直接蒸馏无法训练的问题。

链接：https://arxiv.org/abs/2608.03316

作者：Siming Fu, Zheming Fu, Ruizhe He, Hualiang Wang, Jie Huang, Xiaoxiao Ma, Mingchen Zhong, Weihu Huang, Xiaoxuan He, Haojun Xu

英文摘要：On-policy distillation, in which a teacher corrects samples that the student itself generates, presupposes that the two models speak the same language: identical VAE latents, matching architectures, and a common timestep grid. We ask what happens when none of this holds, as when the strongest teacher available and the student one wishes to deploy come from different model families, and find that the standard recipes have no answer: teacher latents cannot serve as targets in a foreign coordinate system, per-pixel losses against a teacher that stochastically re-draws local detail degenerate into blur or divergence, and timestep indices lose their meaning across mismatched schedules. We present Any-OPD, to our knowledge the first framework for on-policy distillation between arbitrary pairs of latent flow-matching generators. Any-OPD treats the teacher purely as a black-box sampler and connects the two models at exactly one point: a frozen, model-agnostic vision representation in which their independently decoded outputs are compared, sidestepping every assumption about latents, features, or architecture. Trajectory correspondence is recovered by matching continuous noise levels instead of step indices, and a brief anchoring phase, in which teacher samples are re-encoded through the student's own VAE, ensures the on-policy gradient measures sample quality rather than domain mismatch. Distilling the 12B FLUX.1-dev into the 2.5B SD3.5-Medium, Any-OPD lifts the student's PickScore from 0.846 to 0.884 and HPSv3 from 9.12 to 10.97, rivaling the teacher at a fifth of its size, where direct latent regression fails to train at all.

AI 总结：该研究提出可解释混合因果发现框架GENESIS，实现100%决策可追溯性，在多数基准数据集上性能优于纯统计方法，可与先进LLM辅助方法媲美。

链接：https://arxiv.org/abs/2608.03868

作者：Abhinav Thorat, Ravi Kumar Kolla, Vishak K Bhat, Harsh Vardhan Singh Chauhan, Niranjan Pedanekar

英文摘要：Causal Discovery (CD) from observational data faces two fundamental challenges. First, purely statistical methods often lack the power to resolve structural ambiguities in low-sample regimes. Second, although LLM-assisted hybrid approaches improve structure recovery through semantic reasoning, the influence of that reasoning on individual edge decisions remains largely opaque. Consequently, existing hybrid methods fail to satisfy a fundamental requirement: explaining why a particular edge is included or excluded in the learned directed acyclic graph (DAG). This is critical in real-world applications, where no ground-truth DAG exists and every structural decision must be independently justified. We formalize this requirement as decision traceability, requiring every inferred edge to be supported by auditable statistical evidence, Markov Blanket consistency, or explicit domain reasoning. We propose GENESIS, an explainable hybrid CD framework that decomposes graph construction into interpretable decision points. GENESIS first identifies and scores three-node structural motifs, including chains, forks, and colliders, to establish transparent structural priors, then progressively refines the graph by integrating these priors with observational evidence, invoking domain knowledge only when statistical evidence is insufficient. By design, every edge decision is resolved through an auditable source of evidence. Experiments show that GENESIS achieves 100% decision traceability across all settings, establishing explainability as a first-class objective in causal discovery. Despite this additional requirement, GENESIS consistently outperforms purely statistical CD methods on the majority of benchmark datasets across all sample regimes in terms of Structural Hamming Distance (SHD), while achieving performance comparable to state-of-the-art LLM-assisted approaches.

AI 总结：本研究针对合成组织病理学图像生成的评估问题，提出基于数字病理学预训练基础模型改进的FID、IS及精确率-召回率指标，实验发现改进后的IS与下游细胞核分割性能相关性更高，且生成数据多样性对分割性能的提升作用强于单张图像视觉保真度。

链接：https://arxiv.org/abs/2608.03990

作者：Seyed Kahaki, Shijie Li, Weijie Chen, Nicholas Petrick

英文摘要： Synthetic histopathology image generation has emerged as an approach that may address data scarcity in computational pathology, yet current evaluation methodologies may not fully assess synthetic data quality for medical applications. This work investigates and addresses limitations in existing evaluation metrics, investigating an approach for assessing synthetic histopathology image quality through domain-specific metrics and downstream task validation. We show that conventional synthetic data evaluation metrics such as Frechet Inception Distance (FID) and Inception Score (IS) may have limitations when applied to histopathology images due to their reliance on ImageNet-pretrained feature extractors. To address these limitations, we propose for consideration modified FID and IS approaches utilizing foundation models pretrained on digital pathology datasets, supplemented by precision-recall based metrics as part of an additional quality assessment. Using conditional denoising diffusion models trained on four benchmark datasets, with a two-step training approach, we generated synthetic datasets with systematically varied quality characteristics. We also measured the correlation between the synthetic data quality metrics with downstream nuclei segmentation performance using common metrics including the aggregated Jaccard index (AJI+) and the Dice coefficient. The study results suggest that pathology-specific metrics may provide improved discriminative power. Specifically, the modified Inception Score indicates higher correlation with downstream task performance (r=0.6096 with AJI+, p=0.0122), compared to the original IS (r=0.0708, p=0.7944). Our observations indicate that increasing the variety of generated training data has a higher positive correlation with segmentation model performance than improving the visual fidelity of individual generated images.

AI 总结：针对灾后损伤评估的传统方法成本高且难以适应动态条件，本研究提出结合水平集估计的成本感知贝叶斯优化框架，引导自主数据采集器流向高信息区域，经模拟和R2D数据验证可高效评估损伤以支持应急响应。

链接：https://arxiv.org/abs/2608.02868

机构：School of Computing and Augmented Intelligence, Arizona State University（亚利桑那州立大学计算与增强智能学院）； University of Florida（佛罗里达大学）； Stevens Institute of Technology（史蒂文斯理工学院）

作者：Boyang Xu, Mostafa Reisi Gahrooei, Mohammad Ilbeigi, Hao Yan

英文摘要：Natural disasters frequently inflict severe damage to the built environment, which demands a rapid, reliable, and cost-effective damage assessment for emergency response. However, traditional methods for post-disaster damage assessment often rely on static, labor-intensive data collection strategies that can be prohibitively expensive and struggle to adapt to dynamic post-disaster conditions. In this study, we propose a cost-aware Bayesian optimization framework combined with level-set estimation that continuously guides autonomous data collectors, e.g., an unmanned aerial vehicle (UAV), toward the most informative regions. By dynamically updating damage estimates across different geographic zones, our approach systematically reduces uncertainty while minimizing operational costs. The proposed framework is first validated using a controlled synthetic toy study, demonstrating the agent's ability to efficiently trace damage boundaries, recover the underlying damage map, and rapidly reduce predictive uncertainty. Furthermore, the approach is evaluated using high-fidelity disaster data generated by the Regional Resilience Determination (R2D) software. The results of the algorithm provide accurate and timely damage estimates that support informative and fast emergency response.

AI 总结：该研究提出PopFS方法，通过多任务稀疏学习缩小候选池并在硬特征集上搜索，实现群体鲁棒特征选择，在多个数据集的预测任务中提升平均与最差群体性能，且可通过调整福利目标优化特定群体表现。

链接：https://arxiv.org/abs/2608.02887

作者：Ruiqi Lyu, Alistair Turcan, Bryan Wilder

英文摘要：Choosing which features to collect is a deployment decision: the same limited questionnaire, test panel, or sensor set may need to serve several heterogeneous populations. Standard feature-selection methods typically optimize for one large population, while existing robust approaches tend to learn one shared model for every population. We introduce PopFS, a method for learning one shared, deployable feature set that is robust to population differences while letting each pop- ulation train its own model. PopFS uses a tunable welfare objective that lets practitioners balance overall predictive ben- efit against stronger protection of the populations that benefit least. To make this objective practical at scale, PopFS first uses multitask sparse learning to reduce the candidate pool, then searches directly over hard feature sets by ranking promising additions and swaps and fully refitting only a shortlist. Across eight population splits from six prediction tasks drawn from five tabular and public-health datasets, PopFS consistently achieves strong average and worst-population performance while scaling to thousands of candidate features. A 43-state COVID-19 nowcasting study further shows that changing the welfare objective can improve the least-served states with lit- tle change in average performance and yields an interpretable change in the selected symptom signals. Our code is available at this https URL.

AI 总结：该研究针对灰盒优化问题，提出双层贝叶斯优化方法，通过外层优化黑盒变量、内层求解白盒子问题，在13个基准问题上实现更低后悔值，且优势具有鲁棒性。

链接：https://arxiv.org/abs/2608.03045

作者：Joshua E. Hammond, Tyler A. Soderstrom, Brian A. Korgel, Michael Baldea

英文摘要：We consider grey-box optimization problems where the decision variables naturally partition into black-box variables (as arguments to an expensive black-box function) and white-box variables, governed by a set of explicit, closed-form equations that also depend on the output of the black-box function. We exploit this separability through a bilevel reformulation: an outer Bayesian optimization (BO) to optimize the scalar objective as a function of black-box variables alone, while an inner problem solves the white-box subproblem via global optimization. The Gaussian process surrogate used in BO is therefore defined rather than and white-box constraints are satisfied exactly whenever the inner optimizer converges to a feasible point---without penalty functions, chance constraints, or moment approximations. On a suite of 13 benchmark problems, bilevel BO achieves lower regret, with fewer iterations and wall clock time. This advantage is robust to initialization set size, exploration parameters, and inner-solver choice.

AI 总结：本文针对良性插值现象的新解释展开哲学层面的辩论，指出其诉诸单个模型简单性的做法缺乏与泛化的可证关联，造成了解释缺口。

链接：https://arxiv.org/abs/2608.03386

作者：Tom F. Sterkenburg, Daniel A. Herrmann, Jan-Willem Romeijn

英文摘要：Contemporary deep learning methods generalize well even when they fit their training data perfectly, a phenomenon known as benign interpolation. This phenomenon cannot be accounted for by classical statistical learning theory and has prompted a range of attempted new explanations in the statistics and machine learning literature. A common feature of these new proposals is an appeal to a simplicity preference among interpolating models, often presented as a form of Occam's razor. We clarify this debate for a philosophical audience and argue that this new appeal to simplicity creates an explanatory gap. The classical theory offers theorems which connect the simplicity of model classes to good generalization, thus underwriting methodological simplicity norms. The new accounts instead appeal to properties of individual models, which they interpret as a kind of simplicity. Lacking a provable connection to generalization, it is the name "simplicity" that does the work a theorem used to do, making a substantive and unargued assumption look like the application of a familiar methodological principle.

AI 总结：针对长上下文LLM推理的KV缓存内存瓶颈，提出AnchorKV压缩方案，不丢令牌且压缩20倍，70B规模下保留99%全缓存精度，大幅降低上下文成本。

链接：https://arxiv.org/abs/2608.02901

作者：Malik Khalaf, Yara Shamshoum, Nitzan Hodos, Yuval Sieradzki, Assaf Schuster

英文摘要：The key-value (KV) cache is the primary memory bottleneck in long-context LLM inference. Existing approaches attack it from opposite ends: eviction methods permanently discard tokens, degrading performance whenever a discarded token later proves essential, while quantization methods retain all tokens at low precision but offer limited compression. We propose AnchorKV, a compression scheme that shrinks the cache by $20\times$ without discarding a single token. AnchorKV represents the cache using a small set of anchors stored exactly, expresses every other token through its most similar anchor, and refines only those whose approximation most affects the model's output. AnchorKV consistently preserves accuracy across models and datasets, retaining 99% of the full-cache score at the 70B scale, while keeping the entire context at a fraction of its cost.

AI 总结：本研究提出结合硬件感知能量预测模型与多目标优化的方法，在设计阶段离线优化微控制器上的DNN，通过Cortex-M4 MCU验证了方法有效性，实现边缘端自主AI。

链接：https://arxiv.org/abs/2608.03589

作者：Jakob Schubert, Maximilian Kasper, Maximilian Linke, Benedict Herzog, Mark Deutel, Axel Plinge, Dominik Seuss, Christopher Mutschler

英文摘要：We present a method for designing deep neural networks (DNNs) for intermittent, energy-autonomous, on-device learning on microcontroller units (MCUs). In mobile applications where the energy can run out, e.g., when solar-powered, executing artificial intelligence (AI) faces a technical issue as learning can be interrupted at any time. Our approach combines a hardware-aware energy prediction model with multi-objective optimization (MOO), enabling offline DNN optimization at the design stage without repeated deployment and online testing on the target MCU. Our proposed energy predictor estimates per-layer energy consumption for both DNN inference and training, including the intermittent checkpointing overhead, based on implementation-specific compute and memory features extracted from the DNN model. We validate our approach using autoencoders for anomaly detection on a Cortex-M4 MCU, where our predictor achieves a weighted absolute percentage error of 16.6%, which is sufficient for reliable architecture selection under intermittency constraints. As a result, this work bridges the gap between MOO, automated DNN design, deployment on energy-harvesting systems, and intermittent learning, truly enabling autonomous AI at the edge.

AI 总结：本研究评估了面向分词电子健康记录的联邦生成事件模型（GEMs），其跨站点性能优于LightGBM，联邦学习（FedAvg、FedAvgM）表现接近集中式训练，为解决电子健康记录模型的数据孤岛问题提供了可行方案。

链接：https://arxiv.org/abs/2608.02939

机构：University of Chicago（芝加哥大学）； Northwestern University Feinberg School of Medicine（西北大学费恩伯格医学院）

作者：Michael C. Burkhart, Luke Solo, Inhyeok Lee, S'Khaja Charles, Zewei "Whiskey" Liao, Kaveri Chhikara, Dema Therese, Wan-Ting Liao, Catherine A. Gao, William F. Parker, Brett K. Beaulieu-Jones

英文摘要：Electronic health record foundation models are limited by institutionally siloed data and substantial performance degradation under cross-site transfer. We evaluated federated training of tokenized generative event models (GEMs) across 122,251 intensive care hospitalizations from three independent health systems harmonized to the Common Longitudinal ICU Data Format. Models were assessed on 12 post-24-hour clinical prediction tasks using within-site, cross-site, centralized, and federated training configurations. GEMs achieved the highest mean within-site and cross-site ROC-AUC and were substantially more transportable than conventional supervised models: their average cross-site penalties were 0.025 ROC-AUC and 0.027 PR-AUC, compared with 0.079 and 0.089 for LightGBM. Federated Learning (FedAvg and FedAvgM) approached the performance of centralized GEM training, with most gains obtained within 5-10 communication rounds. However, centralized multi-site training provided only modest improvements over complete local training. Multi-site models were most useful when local training data were limited, with their advantage narrowing as institutional data accumulated. These findings show that federated GEM training is technically feasible and preserves most centralized performance, but that the main open challenge is learning transportable representations to translate larger, but heterogeneous data from multiple health systems into a reliable target-site benefit.

AI 总结：针对持续大语言模型遗忘的两大挑战，提出TFR-Net，通过跟踪通道级风险抑制持久目标相关通道、恢复休眠通道，在四个数据集上取得更优的遗忘效果与保留性能权衡。

链接：https://arxiv.org/abs/2608.03123

作者：Zezheng Wu, Xinghe Cheng, Qinggang Zhang, Haoran Luo, Jiapu Wang, Qing Yang, Jingwei Zhang

英文摘要：Machine unlearning aims to eliminate the influence of sensitive data on a model. In the real world, unlearning requests arrive continually, which gives rise to two challenges. First, an unlearning intervention may redistribute target-related computation across remaining pathways, allowing previously forgotten knowledge to re-emerge. Second, repeated unlearning interventions may progressively reduce the model capacity needed to preserve retained utility. To address these challenges, we propose the Trajectory-guided Forget-Recover Network (TFR-Net). TFR-Net tracks channel-level risk across requests. It separates persistent target-related channels from transient hotspots and suppresses only the persistent ones. TFR-Net also recovers model capacity by reactivating dormant channels. These channels make strong contributions to retained utility and show low current and historical forget risk. The recovery is accepted only when retained-utility degradation remains within a predefined tolerance. Experiments on four datasets show that TFR-Net consistently achieves a more favorable trade-off between unlearning effectiveness and retained utility than representative baselines.

AI 总结：本文针对差分隐私零阶微调大语言模型的噪声问题，提出SAGE噪声感知收缩方法，经理论分析和多模型实验验证，该方法在相同隐私预算下多数场景优于基线，且保留内存效率。

链接：https://arxiv.org/abs/2608.03277

作者：Lele Zheng, Weifeng Kong, Xinyi Zhang, Ke Cheng, Tao Zhang, Yulong Shen

英文摘要：Differentially private zeroth-order optimization (DP-ZO) enables memory-efficient private fine-tuning of large language models using only forward evaluations. Existing aggregation-based DP-ZO methods reconstruct model updates at a fixed scale, ignoring that the strength of useful signals varies throughout training. Consequently, noise-dominated updates may receive excessive weight and degrade model utility. To address this issue, we propose SAGE, a noise-aware shrinkage method that adaptively attenuates privatized estimates according to their estimated signal quality. SAGE subtracts the known Gaussian noise variance from the observed second moment to estimate the underlying signal energy, stabilizes this estimate through temporal tracking, and compares its current signal-to-noise level with a warm-up reference to derive a bounded shrinkage factor. As pure post-processing, SAGE requires neither additional privacy budget nor model queries and introduces only constant additional state. Our theoretical analysis shows that shrinkage reduces the quadratic update-risk term faster than the linear descent term, preserving useful descent while limiting the influence of noise-dominated updates. Experiments on RoBERTa-large, OPT-1.3B, and OPT-6.7B demonstrate that SAGE outperforms existing baselines in most settings under the same privacy budgets while preserving the forward-only memory efficiency of DP-ZO.

AI 总结：针对混合ANN-SNN联邦学习的表征失配问题，提出AS-FedBridge框架，通过伪脉冲接口实现信号对齐，在四个数据集上缓解异构挑战并实现性能与效率的可控权衡。

链接：https://arxiv.org/abs/2608.03324

作者：Shengyang Li, Yiting Dong, Liuyang Song, Ximing Wang, Luyuan Xie, Cong Li, Qingni Shen, Zhaofei Yu

英文摘要：Federated learning enables collaborative model training across distributed edge devices while strictly preserving data privacy. To facilitate practical deployment on resource-constrained edge devices, Spiking Neural Networks (SNNs) have emerged as a promising alternative to traditional Artificial Neural Networks (ANNs) due to their sparse computing mechanisms and high energy efficiency. However, jointly training ANNs and SNNs exposes a challenge of representational misalignment, which is intrinsically caused by differences in information representation, specifically the semantic gap between continuous real-valued activations in ANNs and discrete spatio-temporal spikes in SNNs. To overcome this barrier, we propose AS-FedBridge, a novel federated learning framework tailored for mixed ANN-SNN clients. AS-FedBridge features a lightweight Bridge equipped with a Pseudo-Spike Interface, which effectively projects continuous signals into a spike-compatible space to facilitate ANN-SNN alignment. Given the absence of existing mixed ANN-SNN federated frameworks, we establish a comprehensive benchmark to evaluate against multiple advanced heterogeneous FL methods. Our empirical analysis demonstrates a positive correlation between the degree of ANN-SNN alignment and the collaborative FL performance. Across four datasets, AS-FedBridge consistently demonstrates advanced accuracy while mitigating extreme scale, architecture, and client heterogeneity challenge. Furthermore, our framework enables a highly controllable trade-off between model performance and resource efficiency. AS-FedBridge accomplishes these robust performance gains while introducing only marginal computational overhead, establishing a robust and practical foundation for mixed ANN-SNN federated learning systems.

AI 总结：针对医疗联邦的非IID数据、异构目标与私有特征问题，提出FedCARE框架，通过两阶段训练策略实现多目标个性化，在MIMIC-III等数据集上较FedAvg取得显著性能提升。

链接：https://arxiv.org/abs/2608.03498

作者：Rojalini Tripathy, Padmalochan Bera, Shreya Ghosh, Rajkumar Buyya

英文摘要： Federated Learning (FL) enables collaborative model training across distributed healthcare institutions without centralising sensitive patient data. However, real-world healthcare federations are often characterised not only by non-IID data, but also by heterogeneous clinical objectives and partially overlapping feature spaces. Different hospitals may optimise distinct and potentially conflicting objectives, such as mortality risk prediction, readmission reduction, or length-of-stay estimation, while also retaining institution-specific clinical features that cannot be shared with other participants. Existing personalised FL methods mainly address statistical heterogeneity, whereas multi-objective FL approaches typically learn a shared global model without explicit client-level adaptation. To address these limitations, we propose \textbf{FedCARE}, a multi-objective personalised FL framework for smart healthcare services. FedCARE follows a two-stage training strategy. First, it learns a shared global backbone from common clinical features using Pareto-driven multi-objective federated optimisation. Second, each client independently fine-tunes the shared backbone using its private features and local clinical objectives, enabling institution-specific personalisation without additional communication overhead. We implement FedCARE in a cloud-based client-server federated deployment on the Melbourne Research Cloud and evaluate it on two real-world healthcare datasets, MIMIC-III and Diabetes 130-US Hospitals. Experimental results show that FedCARE consistently outperforms standard FL, multi-objective FL, and personalised FL baselines, achieving up to 12.5% AUROC improvement and 32.0% MAE reduction over FedAvg.

AI 总结：针对叠层成像神经网络分布外泛化时的缩放不一致问题，提出分解策略与合成物体采样策略，在5个实验数据集上将傅里叶误差较PtychoPINN-torch基线降低最多5倍，提升了其实际应用可行性。

链接：https://arxiv.org/abs/2608.02869

机构：Argonne National Laboratory（阿贡国家实验室）； SLAC National Accelerator Laboratory（SLAC国家加速器实验室）； Lawrence Berkeley National Laboratory（劳伦斯伯克利国家实验室）

作者：Albert Vong, Steven Henke, Oliver Hoidn, Hanna Ruth, Junjing Deng, Apurva Mehta, David Shapiro, Alexander Hexemer, Nicholas Schwarz

英文摘要：Ptychography neural networks suffer from scaling inconsistencies when generalizing out of distribution, limiting their real world viability. We address this scaling mismatch using a factorization strategy which decouples the learned object texture from measurement scaling, enabling a single trained network to produce measurement-consistent reconstructions across varying illumination conditions. This requires predicting the learned object in real and imaginary units instead of the canonical amplitude and phase representation. We additionally introduce a synthetic object sampling strategy that minimizes phase distribution mismatch between synthetic training data and experimental targets. These improvements yield up to a 5x reduction in Fourier error over the previous PtychoPINN-torch baseline across 5 experimental datasets spanning multiple beamlines and facilities.

AI 总结：该研究针对 Mistral-7B 变体在 PubMed RCT 句子分类任务上，评估量化及提示模板、评分规则等对生物医学大语言模型可靠性的影响，发现这些因素对校准和准确率的影响不可忽视。

链接：https://arxiv.org/abs/2608.03854

作者：Anton Rasmussen, Hong Qin

英文摘要：When decoder language models are used as classifiers, predicted class probabilities depend on implementation choices, including the prompt template, verbalizer (label-to-token mapping), and scoring rule, that are rarely treated as experimental variables. We present a controlled evaluation of three Mistral-7B variants (Base, BioMistral, and Instruct) on PubMed RCT sentence classification (n=2000) under FP16, INT8, and INT4 precision using four answer-text prompt templates. Our primary finding is that the probability extraction protocol dominates apparent calibration. Switching from summed to mean token log-likelihood scoring reverses the calibration ranking between models: BioMistral average expected calibration error increases from 0.097 to 0.289, whereas Instruct decreases from 0.237 to 0.096, while accuracy changes by less than 1 percentage point for the specialized models but 4-6 percentage points for the base model. Prompt template choice produces accuracy differences of 7-24 percentage points, comparable to or larger than model-level effects. On one template, BioMistral outperforms Instruct although the overall mean favors Instruct by only 1.3 percentage points. For BioMistral and Instruct, INT8 quantization changes accuracy and F1 by only 1-2 percentage points relative to FP16, whereas the base model shows larger INT8 effects on some templates (up to +4.2 percentage points). INT4 produces heterogeneous but non-catastrophic effects. Temperature scaling reduces expected calibration error under summed scoring for both models but only for that scoring rule. A fine-tuned PubMedBERT reference achieves 82.7% accuracy but uses about 176000 labeled training examples, precluding direct comparison. These results demonstrate that prompt template design and scoring normalization are first-order experimental decisions when evaluating decoder language model calibration.

AI 总结：该研究针对时序图聚类任务，从原理、基元与池化三方面展开，提出 GPU 加速基元实现高效时序聚类，为理论导向的池化提供路径，明确了算法与神经模型的适用场景。

链接：https://arxiv.org/abs/2608.03696

作者：Nelson Aloysio Reis de Almeida Passos, Emanuele Carlini, Salvatore Trani

英文摘要： This work focuses on the problem of learning on temporal graphs, with particular emphasis on the task of clustering: obtaining coarse-grained representations by aggregating information from nodes, edges, and temporal dynamics - a task related to pooling in machine learning on graphs, or community detection in network science. Although graph neural networks reach state-of-the-art performance across many downstream graph tasks, their advantage over established descriptive and inferential clustering algorithms is far less settled, especially under demands of efficiency and recovery accuracy. We frame this tension through three linked perspectives: principles, connecting graph learning and community detection through shared spectral foundations and detectability thresholds in stochastic block model regimes; primitives, making spectral clustering and multislice modularity optimization tractable through GPU-accelerated temporal backends; and pooling, viewing principled community detection as a theory-grounded coarse-graining operator for temporal graphs. Our results indicate that algorithmic methods remain the appropriate tool where attributes are absent or weak - scalability rather than accuracy being the binding obstacle - while neural models are most compelling when structural, temporal, and attribute signals align. By making temporal clustering scalable, GPU-accelerated primitives suggest a route toward theory-grounded pooling, while raising a central question: when does community-based coarse-graining preserve the dynamics needed for downstream learning tasks?

AI 总结：

链接：https://arxiv.org/abs/2608.03579

作者：Xiang Li, Pengcheng Wang, Huazheng Wang, Saurabh Bagchi

英文摘要：Modern multi-tenant Low-Rank Adapters (LoRAs) serving systems concurrently host tens to hundreds of LoRA adapters. Though powerful, this introduces a critical system dilemma between serving efficiency and task performance: higher-rank adapters generally achieve better downstream task performance, but their GPU VRAM footprint and Host-to-Device PCIe swapping overhead severely constrain scalability. Conversely, ultra-low-rank adapters ($r \le 2$) minimize both VRAM footprint and PCIe transfer overhead, but suffer from downstream task performance degradation. To solve this problem, we propose Subspace-Aligned LoRA Training (SALT), a serving efficiency-aware hierarchical fine-tuning framework. Our solution operates in three phases. First, a provider jointly trains high-capacity domain centroids on public data within the domain using a novel alignment regularizer that coheres in-domain task subspaces into a unified basis. Next, users fine-tune ultra-low-rank task residual adapters on private data atop those frozen centroids. Finally, during inference, the provider pins the centroid in GPU VRAM and dynamically swaps in each user's task residual on demand. Across LLMs of varying scales, SALT recovers high-rank accuracy using $r \le 2$ residuals, achieving up to 18.5% absolute accuracy gains over state-of-the-art compression baselines and reducing per-adapter memory by up to 16x. When integrated into vLLM, SALT improves serving throughput by up to 51% under PCIe bandwidth pressure and 28% under GPU VRAM constraints for Llama-3.2-3B.

AI 总结：Omega-S是一种仅依赖权重矩阵的即插即用惩罚项，在Llama-3-8B的LoRA微调中，它在保留模型原有能力上优于无正则化、调优后的权重衰减和EWC，且成本增加不到4%。

链接：https://arxiv.org/abs/2608.03887

作者：Alberto Acedo

英文摘要：Fine-tuning a large language model on new data degrades what it previously learned. We present Omega-S, a drop-in penalty computed from the weight matrix alone: it needs no previous-task data, no Fisher matrix and no stored copy of the old weights. It is three lines in an existing training loop and adds under 4% to the cost of a step. Retention. On Llama-3-8B with LoRA, fine-tuned from code to prose and measured by HumanEval over ten seeds, Omega-S retains more of the original capability than no regularisation on 9 of 10 seeds (0.173 -> 0.238 absolute pass@1; sign test one-sided p=0.011, Wilcoxon p=0.006), as a retention ratio, 62.9% -> 84.1%. It also beats tuned weight decay on 10 of 10 seeds (p=0.002) and tuned EWC on 8 of 10 (p=0.014), every arm re-measured in the same session. Mechanism, measured rather than asserted. Omega-S is topological by construction, its objective built from Tr(A^3), but we measured which of its four factors actually moves and three do not: their elasticity with respect to the weights is at or below 1e-4, against 9e-3 for the degree-variance term. As implemented, the composite reduces to a penalty on the variance of node degrees, which means row magnitude in square modules and directional alignment in non-square ones. We report this because a method whose name promises one thing and whose gradient does another should say so. We also enumerate the open design choices, including a contrast-preserving construction that does what it was designed to do and makes retention worse on all ten seeds. Repeating an identical configuration, same seed and same hardware, gives a standard deviation of 0.104 in retention ratio. We have not found this quantified for low-rank fine-tuning of language models, and it bounds every seed-paired comparison in this literature, ours included. Code, per-seed results and the full record of negative results are available.

AI 总结：该研究提出基于q-厄米特I多项式的新型q正交核，其在20个基准数据集上的性能与经典核及其他正交多项式核相当，且数值稳定性和计算简便性更优，为核设计提供了新方向。

链接：https://arxiv.org/abs/2608.03482

作者：Álvaro Sánchez-Paniagua Ríos, Juan P. Llerena, Alberto Lastra, Nuria Torrado, Edmundo J. Huertas

英文摘要：The performance of Support Vector Machines (SVMs) critically depends on the kernel function choice, which enables implicit mapping of data into high-dimensional feature spaces. While classical kernels like Radial Basis Function (RBF) remain popular, orthogonal polynomial kernels offer mathematically interpretable alternatives that can incorporate structured prior knowledge. This work extends the orthogonal polynomial kernel paradigm by introducing a novel family based on discrete $q$-Hermite I polynomials, a class of $q$-orthogonal polynomials that generalize classical Hermite polynomials through a deformation parameter $q$. We formally define the q-Hermite kernel and establish its validity under Mercer's theorem. The kernel's inherent boundedness properties naturally prevent annihilation and explosion effects without requiring explicit scaling mechanisms. Extensive experiments across 20 benchmark datasets demonstrate that the proposed kernel achieves competitive performance compared to both classical kernels and other orthogonal polynomial kernels, while offering advantages in numerical stability and computational simplicity. Our results confirm that $q$-orthogonal polynomials constitute a promising direction for kernel design, bridging mathematical elegance with practical machine learning applications, that provides conceptual and algorithmic resources that may be further extended to emerging quantum computing paradigms. To facilitate full reproducibility, we provide the complete implementation and experimental pipeline in an open-access GitHub repository at this https URL.

AI 总结： 本文提出感知可行性的分层扩散分布学习框架，学习电网拓扑、参数与负荷的交流可操作联合分布，生成的合成电网场景可行性与鲁棒性显著提升，且消除了优化后处理。

链接：https://arxiv.org/abs/2608.03878

作者：Chenhan Xiao, Xinyu He, Haoran Li, Hanghang Tong, Yang Weng

英文摘要：Synthetic power-grid scenarios are essential for planning, resilience assessment, contingency analysis, and data-driven power-system applications. Recent synthetic grid generation methods have improved structural realism and operational feasibility by incorporating engineering knowledge through post-generation validation, optimization, or physics-aware generation. However, generated scenarios may still exhibit low AC feasibility and robustness, limiting their practical value for downstream power-system studies. This paper proposes a feasibility-aware distribution-learning framework that learns the AC-operable joint distribution of network topology, branch electrical parameters, and time-varying load profiles. Instead of enforcing feasibility after generation, the proposed framework incorporates AC power-flow convergence and operational constraints into hierarchical diffusion-based distribution learning. This enables the generator itself to produce operationally feasible grid scenarios through efficient diffusion sampling. The hierarchical architecture decomposes the high-dimensional generation task into three engineering-motivated stages: topology and bus-attribute generation, branch-parameter generation conditioned on the generated structure, and load-profile generation conditioned on both network structure and electrical characteristics. Experiments on benchmark systems demonstrate that the proposed framework significantly improves operational feasibility and contingency robustness while maintaining strong statistical fidelity and eliminating optimization-based post-processing.

AI 总结：本研究提出GeoID-PINN，一种结合地理耦合与正则化的物理信息神经网络，用于区域流行病SIRD动力学推断，在模拟数据与路易斯安那州64个县COVID-19数据上均提升了预测准确性。

链接：https://arxiv.org/abs/2608.02633

机构：University of Michigan（密歇根大学）

作者：Weixiong Hua, Fan Bu

英文摘要：Regional surveillance data reflect local transmission, reporting, seeding, and external infection pressure, which are difficult to identify separately. We introduce GeoID-PINN, a physics-informed neural network (PINN) for susceptible-infectious-recovered-deceased (SIRD) dynamics. The model represents spatial dependence with a row-stochastic source-composition matrix whose rows assign nonnegative source weights that sum to one. We regularize this matrix toward a spatial prior constructed from distance, adjacency, commuting, or lead-lag information. In a four-region simulation with known truth, a compatible distance prior gives source-composition error 0.099. The error rises to 0.159 without regularization and 0.577 under a strongly misspecified prior, while trajectory fit and transmission-scale estimates remain similar. Accurate trajectories therefore do not guarantee recovery of the regional dependence structure. We also evaluate GeoID-PINN retrospectively using COVID-19 data from 64 Louisiana counties. Relative to an autoregressive negative-binomial baseline, Forecast-Trained Geo-PINN reduces mean squared error (MSE) from 32,957 to 11,468 and mean absolute error (MAE) from 70.60 to 57.73. The baseline has lower negative log likelihood (NLL), 5.158 versus 5.346, indicating better distributional fit but worse point accuracy. In a controlled 15-county comparison, county adjacency reduces MSE by 6.85 percent and MAE by 3.1 percent. Similar performance across plausible priors supports structured regularization but not unique edge recovery. These results require prior-sensitivity and observation-model checks before interpretation.

AI 总结：该研究针对收入预测无法拆解驱动因素的问题，开发CBMT模型，利用966家公司数据验证其准确性优于多数基准，发现客户群协同变动强时联合预测更有效，同时指出高波动下模型优势缩小。

链接：https://arxiv.org/abs/2608.02911

机构：University of Wisconsin–Madison（威斯康星大学麦迪逊分校）； University of Maryland（马里兰大学）； Boston University（波士顿大学）

作者：Kyeongbin Kim, Daniel McCarthy, Dokyun Lee

英文摘要：Revenue forecasts guide acquisition budgets, demand planning, and customer-based valuations, yet an aggregate forecast does not show whether change reflects acquisition, repeat purchasing, spending per order, or offsetting movements. Using weekly transaction panels for 966 companies in 25 industries, the authors develop the Customer-Based Multi-task Transformer (CBMT), which learns shared structure, retains separate primitive forecasts, and aligns their combination with downstream revenue. CBMT's mean total-sales error is 30% below the strongest representative established customer-base benchmark. It is also 2.65% below a Transformer that forecasts total sales directly, although the paired difference is not statistically significant (p=.222), and it beats separately estimated single-task forecasts for 74.3% of firms. CBMT's source MAE is lower in 23 of 24 benchmark-by-outcome comparisons, with the remaining difference not statistically distinguishable from zero. Firms whose primitives co-move more strongly are more likely to benefit from joint forecasting; selected-family scenario-3 comparisons are consistent with gains from shared representation and revenue alignment but remain diagnostic rather than causal. Accuracy deteriorates for all models when customer-base dynamics are highly volatile, and CBMT's advantage narrows there. Calibration-period routing rules do not improve average accuracy over always deploying CBMT. The results show how coordinated customer-base forecasts support revenue planning and when they warrant greater caution.

AI 总结：本研究开发了将航班时刻表转换为安检负荷信号的框架，结合时间融合Transformer模型，在机场每小时吞吐量预测中优于RNN和LSTM，可支撑安检点人员配置与规划。

链接：https://arxiv.org/abs/2608.02950

机构：School of Aviation and Transportation Technology, Purdue University（普渡大学航空与运输技术学院）； Department of Geography, The Ohio State University（俄亥俄州立大学地理系）； College of Aeronautics and Engineering, Kent State University（肯特州立大学航空与工程学院）

作者：Yinxiao Zhang, Sen Wang, Yi Gao

英文摘要：Checkpoint staffing requires accurate forecasts of when screening demand will occur, yet flight schedules record departure times rather than passenger arrival times at security checkpoints. This study develops a framework that converts known flight schedules into temporally aligned signals for forecasting hourly checkpoint throughput. Using 2023-2024 Transportation Security Administration throughput data and Cirium Diio flight schedules for Hartsfield-Jackson Atlanta International Airport, domestic and international seat capacity was distributed across pre-departure hours using truncated Poisson kernels. A Temporal Fusion Transformer then combined these schedule-derived arrival-intensity signals with historical throughput, scheduled activity, and temporal variables. Models were trained chronologically, with July-December 2024 reserved for testing, and evaluated against recurrent neural network and long short-term memory models across five random seeds. For direct six-hour forecasts, the proposed model achieved a weighted mean absolute percentage error of 9.33%, compared with 12.16% for the recurrent neural network and 11.37% for long short-term memory, while also producing the lowest errors during peak periods. With six-hour recursive updates, errors remained between 10.60% and 11.04% across 24-96 hour horizons, although longer horizons contained fewer valid forecast origins. By transforming scheduled departures into interpretable pre-departure screening-load signals without requiring passenger-flight matching, the framework supports advance staffing, lane-opening, and multiday checkpoint planning. Because observed throughput reflects realized processing rather than unconstrained arrivals, the forecasts should be interpreted together with local staffing, capacity, queue, and wait-time information.

AI 总结：针对功率磁器件瞬态磁化预测难题，提出物理信息混合神经算子PI-HNO，结合局部循环分支与类Preisach全局分支，在MagNetX数据库14种铁氧体材料上实现了高精度紧凑模型，验证了各组件的独特贡献。

链接：https://arxiv.org/abs/2608.02965

机构：School of Electrical and Data Engineering, the University of Technology Sydney（悉尼科技大学电气与数据工程学院）； School of Electrical and Computer Engineering, the University of Sydney（悉尼大学电气与计算机工程学院）； National Railway Research and Design Institute of Signal and Communication（中国铁路通信信号集团公司通信信号研究设计院）

作者：Yachao Zhu, Qiujie Huang, Sinan Li, Yang Li, Gang Lei, Jianguo Zhu

英文摘要：Magnetic components in high-frequency, high-power-density converters are increasingly driven by non-sinusoidal flux-density waveforms with fast transitions, minor-loop operation, dc bias, and temperature variation. Under these conditions, steady-state core-loss formulas and single-valued material curves cannot fully capture transient magnetization responses. This work proposes the Physics-Informed Hybrid Neural Operator (PI-HNO), a compact material-specific neural model with B-H energy-consistency regularization for core-loss-oriented transient magnetization prediction. Given the measured B(t)-H(t) history, the input B(t) series over the prediction interval and operating-condition information, PI-HNO predicts the H(t) series and the corresponding reconstructed B-H trajectory. The model integrates a local recurrent branch for boundary-state representation and rate-dependent response evolution with a Preisach-inspired global branch that extracts waveform-level hysteresis context. Evaluation on the MagNetX transient database using material-specific models for 14 ferrite materials demonstrates that PI-HNO achieves a compact trade-off between sequence accuracy and B(t)-H(t) energy consistency, with the mean and 95th percentile B(t)-H(t) energy consistency errors of 1.92% and 7.60%, respectively, using only 4777 trainable parameters per model. Ablation studies further demonstrate that the local, global, and energy-aware regularized components provide distinct contributions to transient magnetization prediction.

AI 总结：本研究推出金融时间序列预测基准FinVerse，其针对43个公开时间序列基础模型的分析显示，通用预测标准下的优异表现未必对应实用金融预测，凸显了领域感知基准的必要性。

链接：https://arxiv.org/abs/2608.03259

作者：Jaehoon Lee, Jun Seo, Seunghan Lee, Tae Yoon Lim, Dongwan Kang, Hwanil Choi, Minjae Kim, Sungdong Yoo, Junhyeok Kang, Sangjun Han, Soonyoung Lee, Wonbin Ahn

英文摘要：As time-series foundation models have emerged, the need for benchmarks that can evaluate their forecasting ability in meaningful ways has become increasingly important. Existing time-series forecasting benchmarks provide useful standardized comparisons, but they often evaluate heterogeneous series with uniform error-based metrics. Strong performance under such metrics does not necessarily imply that a model's forecasts will support the best real-world decisions across domains. For example, in stock forecasting, correctly predicting whether a price will rise or fall can be more directly relevant to realized returns than minimizing point-wise forecast error alone. To this end, we introduce FinVerse, a finance-domain time-series forecasting benchmark that takes a first step toward more realistic evaluation. The released FinVerse data artifact contains 116,897 financial time series with 171.1M observations, of which 60,232 series with 17.4M observations are selected as evaluated targets based on their economic relevance to financial decisions. Unlike generic forecasting benchmarks that primarily emphasize uniform point-forecast or probabilistic accuracy, FinVerse defines 11 metric families comprising 78 evaluation metrics and assigns the most appropriate evaluation metrics to each individual time series based on its underlying economic meaning. Our analysis of 43 public time-series forecasting foundation models shows that strong performance under generic forecasting criteria does not necessarily translate into useful financial forecasts. This finding highlights the need for domain-aware benchmarks that evaluate models under objectives closer to real-world decision making.

AI 总结：针对周期性漂移下时间序列预测的相位适配难题，提出基于SO(2)特征旋转的POEM框架，结合DPIA实现相位校正，实验验证其性能竞争力与轨迹规整效果。

链接：https://arxiv.org/abs/2608.03630

作者：Jiawen Zhu, Shuhan Liu, Shengxuan Li, Qiming Shi, Di Weng

英文摘要： Deep learning has advanced time series forecasting, but periodicity drift, in which cycle timing and phase vary over time, remains a challenging problem. Existing methods predominantly model these sequences on fixed time grids, suffering from a limited ability to accommodate phase-related variation. To address this limitation, we propose \textbf{POEM}, a phase-aware forecasting framework based on latent feature rotation using the special orthogonal group in two dimensions, denoted by $\mathrm{SO}(2)$. POEM aims to reduce the phase-related variability by learning a phase-correction coordinate and applying an invertible $\mathrm{SO}(2)$-based rotation to paired latent features. To extrapolate this correction coordinate, Directional Phase Increment Attention (DPIA) retrieves historical phase increments from similar temporal contexts and integrates them into future phase corrections. Experiments demonstrate that POEM achieves competitive performance, while qualitative visualizations suggest that the learned phase-aware transformation makes latent trajectories more regular.

AI 总结：本文针对多元CIR过程，提出摊销干预预测框架CIR-ACTIVA，可估计因果效应、预测多horizon冲击响应，在CDS利差测试中优于基线，能解答观测型方法无法处理的假设性查询。

链接：https://arxiv.org/abs/2608.03715

作者：Andreas Sauter, Sumit Sourabh, Drona Kandhai, Erman Acar

英文摘要：Mean-reverting dynamics are pervasive in finance, and the Cox--Ingersoll--Ross (CIR) process is a standard model for the time series they produce, from short rates to credit default swap (CDS) spreads. Yet CIR models capture only \emph{correlated} co-movement, not \emph{causal} influence between series, so they cannot answer the system's response when one series is externally shocked, which observational conditionals confound with historical co-movement. We make two contributions. First, an amortized model for distributional causal effect estimation that frames trajectories as time-stamped observations and predicts the calibrated multi-horizon shock response without retraining per scenario. Second, a causal multivariate CIR data-generating process that supplies the paired observational and interventional ground truth that real markets cannot. We instantiate and calibrate the framework on CDS spreads as a testbed. CIR-ACTIVA's validity is established on synthetic ground truth, independent of how well the simulator matches reality, while practical grounding is assessed by backtesting the generated traces against real CDS data. Against observational and amortized causal-inference baselines, CIR-ACTIVA leads on both causal selectivity in the joint distribution and horizon-resolved calibration, retaining its selectivity once the interventional law varies over the horizon, with gains concentrating at short horizons. This opens up a class of what-if queries on coupled spread systems, CDS stress testing among them, that observational forecasters cannot answer.

AI 总结：PRISM是用于多元异常检测的即插即用元工作流，通过构建基于图像的时间序列表示，在14个数据集中的10个上取得最佳VUS-PR，冻结ImageNet预训练编码器可高效迁移至TSAD，性能损失小且训练更快。

链接：https://arxiv.org/abs/2608.03926

作者：Mateusz Smendowski, Kamil Faber, Piotr Nawrocki, Nathalie Japkowicz, Roberto Corizzo

英文摘要：Time series anomaly detection (TSAD) underpins applications in predictive maintenance, finance, and cloud computing, however performance remains sensitive to representation choices, especially in multivariate settings. While transforming time series into images has shown success in forecasting and classification, it remains unclear how multivariate, high-dimensional series should be mapped to multi-channel images and whether vision backbones can match time-domain baselines in TSAD. We introduce PRISM, a plug-and-play meta-workflow enabling systematic construction and evaluation of image-based representations for multivariate TSAD. Our evaluation spanning over 7,000 experiments shows that well-designed PRISM configurations are competitive with 24 time-domain baselines, achieving the best VUS-PR on 10 of 14 datasets, with an average improvement of 41% over the best competing method on those datasets. Further, we identify channelization - how the channel dimension of multi-channel images is constructed - as a critical and previously understudied design dimension, and introduce MSM, a novel statistics-based scheme achieving 11-27% gains over PCA-based alternatives. Finally, ImageNet-pretrained encoders transfer effectively to TSAD, with frozen encoders retaining 92% of fine-tuned performance while training 1.8 times faster. Our code is available at: this https URL.

AI 总结：针对现有符号回归方法的缺陷，提出DDRSR方法，拓宽表达式分解简化适用范围、规避暴力搜索，在相关任务中展现显著优势，明确适用场景与未来方向。

链接：https://arxiv.org/abs/2608.02628

作者：Yusong Deng, Yanjie Li, Weijun Li

英文摘要：Symbolic regression (SR) is the task of discovering underlying patterns from data and representing them using mathematical expressions. Current machine learning approaches to SR often lack a profound understanding of the intrinsic mathematical and physical principles governing these expressions. While the pioneering AI Feynman method leverages the mathematical properties underlying the data, its expression simplification mechanism suffers from a narrow scope of applicability and is prone to failure on complex equations. Furthermore, its underlying mechanisms rely heavily on brute-force searches for sub-expressions, severely limiting its practical utility. Through rigorous mathematical deduction and proofs, we propose our method, Deep Divide and Reduce in Symbolic Regression (DDRSR). DDRSR fundamentally broadens the applicability of expression decomposition and reduction, circumvents the need for brute-force sub-structure searches, and ensures both wider versatility and strict theoretical correctness. Empirical evaluations demonstrate that these theoretical principles yield significant advantages in both expression decomposition and numerical regression tasks. Finally, we discuss the applicable scenarios and inherent limitations of this paradigm, alongside promising directions for future research.

AI 总结：本研究提出多模态自回归Transformer代理模型，用于地质碳封存中可变操作建模与不确定性量化，经4000次GEOS流模拟训练后，可精准预测相关参数并降低关键元参数的不确定性。

链接：https://arxiv.org/abs/2608.02629

机构：Stanford University（斯坦福大学）

作者：Yifu Han, Louis J. Durlofsky

英文摘要：The use of variable well perforation and injection strategies can improve the efficiency of geological carbon storage operations. We develop a new multimodal auto-regressive transformer surrogate to model these operations under geological uncertainty. A modified SEAM CO2 geomodel, which involves a faulted system with three stacked aquifers, is considered. The two injection wells are perforated in stages, from bottom to top, with the stage durations and individual well injection rates treated as control variables. The surrogate model processes three input modalities - the 3D geomodel, scalar parameters characterizing relative permeability functions, and control variables - through separate encoders. These are fused via self-attention in a transformer encoder, and a temporal decoder generates predictions auto-regressively through encoder-decoder cross-attention. The surrogate is trained, using 4000 GEOS flow simulations, to predict saturation and pressure at monitoring locations, total injected and mobile CO2 mass, and saturation footprints. For a new test set, involving randomly sampled geomodels and control variables, the surrogate achieves a median saturation MAE of 0.028 and median relative errors of 0.2-5% for the other quantities of interest. Importantly, it captures the switch from rate to bottom-hole-pressure control. The surrogate model is used within a hierarchical Markov chain Monte Carlo data assimilation procedure for a synthetic true model under three operational strategies. Substantial uncertainty reduction is achieved for key metaparameters, particularly the fault permeabilities. Posterior predictions for saturation footprints and total injected and mobile CO2 mass are also shown to be generally consistent with true model results.

AI 总结：该研究提出了一种验证器引导式（VG）工作流程，基于预训练符号 Transformer ODEFormer 实现物理动力系统的可解释预测，在范德波尔振荡器和涡旋脱落场景中展现出优于原方法的性能与泛化能力。

链接：https://arxiv.org/abs/2608.02662

机构：Imperial College London（伦敦帝国学院）

作者：Farbod Faraji, Francesco Belardinelli

英文摘要：Reliable forecasting of nonlinear physical systems underpins scientific discovery and engineering decision-making. Yet high-fidelity simulations are prohibitively costly, and machine-learning surrogates can be opaque and encode assumptions about system dynamics, limiting generalizability. Pretrained transformers mapping synthetic ODE trajectories to equations offer interpretable alternatives, promising transfer without system-specific equation knowledge. Transferring them reliably to high-dimensional physical data, however, remains an open challenge. We develop a verifier-guided (VG) workflow around ODEFormer as a symbolic backbone, using dynamical and physical-admissibility criteria to select from a multi-trajectory candidate equation pool, enabling transfer. On canonical Van der Pol oscillators, VG outperforms the original ODEFormer workflow across held-out initial conditions. We then address vortex shedding, a phenomenon occurring in atmospheric and plasma systems of societal relevance, through coordinate reduction and symbolic discovery at fixed and varying Reynolds numbers. VG discovers fixed-parameter reduced-order equations that recover the fundamental shedding oscillator and higher harmonics without a wake-specific candidate library or prescribed Navier-Stokes structure, while the cross-parameter model generalizes to withheld regimes. Reconstruction fidelity alone did not determine symbolic discoverability, highlighting the importance of compatibility between latent dynamics and the backbone's pretraining distribution. This work establishes a verifier-guided neural-to-symbolic methodology for interpretable and physically auditable forecasting in the natural sciences.

AI 总结：该研究针对ICU死亡率预测中不规则临床观测建模问题，提出CT-HEG架构并实例化为CHIRP-Net，经MIMIC-IV v3.1数据集实验，验证了双向连接、时间注意力边特征等架构选择的作用，为不规则电子健康记录数据建模提供了有效方案。

链接：https://arxiv.org/abs/2608.02663

机构：Westcliff University（韦斯特克利夫大学）； Northern University Bangladesh（孟加拉国北方大学）； Stanton University（斯坦顿大学）； Washington University of Science and Technology（华盛顿科技大学）

作者：Mohammad Nasir Uddin, Rahnuma Tabassum Orpita, Asaduzzaman Anik, Eklachur Rahman Bhuiyan, Marjahan Risalat, SM Wali Ullah, Asif Ahamed

英文摘要：Accurate ICU mortality prediction requires modeling irregular clinical observations across heterogeneous entity types. Existing sequence models handle irregular sampling but ignore typed relational structure; existing graph models assume fixed-interval inputs. We introduce the Continuous-Time Heterogeneous EHR Graph (CT-HEG) schema and evaluate which architectural choices drive predictive performance. CT-HEG encodes each ICU stay as a typed, timestamped graph with three node types (visit, vital, lab_event) and 2D edge attributes (t_hours/48, value_norm) encoding timing and value without imputation. We instantiate CT-HEG as CHIRP-Net, a four-layer heterogeneous GATv2Conv network, evaluated on MIMIC-IV v3.1 (31,142 ICU stays, LOS>=48h, 13.4% mortality) with five seeds and bootstrapped confidence intervals, against logistic regression, mTAND, a Transformer, and GRU-D, plus an ablation study. CHIRP-Net achieved 5-seed mean AUROC 0.8449+/-0.0071 (AUPRC 0.4958+/-0.0209); the ensemble achieved AUROC 0.8618 (95% CI: 0.8485-0.8745). Removing reverse edges disconnected observation nodes from the visit readout, cutting AUROC by 0.1968+/-0.0073. Time-attentive edge features contributed 0.0247+/-0.0093 AUROC. Collapsing heterogeneous edge types into one relation (7x fewer parameters) outperformed the full model on all seeds. Post-calibration ECE was 0.0307. Temporal and demographic subgroup analyses were explored but not reported here, pending follow-up work. Bidirectional connectivity was necessary for the model to use its inputs at all, and CT-HEG was reasonably well calibrated after validation-fitted temperature scaling. These results support CT-HEG for irregular EHR data, while external validation, a pre-specified temporal evaluation, and a demographic fairness audit remain necessary before any claim of robustness. Code: this https URL.

AI 总结：该研究提出将残差连接的收缩映射统一为单参数角收缩族，得到Proj-SpheretNorm、Cay-SpheretNorm等方法，在nanoGPT上验证其性能优于现有方案，发现指数映射并非球面残差流的最优选择。

链接：https://arxiv.org/abs/2608.02668

机构：National Central University（中央大学）； National Yang Ming Chiao Tung University（阳明交通大学）

作者：Jie Zhang, Cheng-Fang Su, Yi-Jui Huang, Min-Te Sun

英文摘要： Residual connections are the de facto mechanism for training deep neural networks stably. Geodesic Normalization (GeoNorm) recasts them on a Riemannian manifold, orthogonalizing each layer output against the current hidden state and applying the resulting update through the Riemannian exponential map. Every hidden state thus keeps a constant $\ell_{2}$-norm, confining the residual stream to a hypersphere. The exponential map, however, is only one member of a broad family of retraction maps. We show that on the hypersphere this entire family collapses to a single scalar design choice. What distinguishes one retraction from another is only how the magnitude of an update is converted into a rotation angle within the plane spanned by the hidden state and the update. This view places Euclidean residual connections and GeoNorm in one framework. Instantiating it with the metric projection retraction and the Cayley retraction yields Proj-SpheretNorm and Cay-SpheretNorm, which are exactly norm-preserving yet require only algebraic operations. Both prove to be members of a one-parameter family of angular retractions, $p$-SpheretNorm, whose rotation angle saturates rather than growing without bound. The two methods above are recovered exactly at $p = 1$ and $p = 2$, while the identity map and GeoNorm arise only as limits at either end. On nanoGPT, all three methods outperform existing lightweight deep connection schemes, and the best validation loss is attained at finite $p$, indicating that the exponential map is not the preferred retraction for spherical residual streams but merely one end of a spectrum.

AI 总结：本研究提出名为GLOBE的轨迹对齐核心集选择框架，通过多阶分布匹配与结构化稀疏优化提升核心集质量，在低保留率下的下游测试准确率优于现有方法，助力设备端数据高效学习。

链接：https://arxiv.org/abs/2608.02690

作者：Hetian Liu, Jin Cui, Mengcheng Shi, Yanbin Hu, Xinyue Long, Boran Zhao, Pengju Pen

英文摘要：On-device training of deep neural networks is fundamentally constrained by the computational and memory costs of large-scale datasets. Coreset selection offers a practical solution by retaining only a compact subset of real training samples. However, existing gradient-based methods commonly rely on gradients computed at a single model snapshot and employ greedy or pursuit-based selection procedures, limiting their ability to capture evolving optimization dynamics and handle strongly correlated samples. We propose GLOBE (Gradient Local-Balanced Extraction), a trajectory-aligned coreset selection framework that formulates sample selection as a globally optimized sparse weighting problem. GLOBE represents each sample by a gradient trajectory constructed across multiple training checkpoints, thereby capturing its influence throughout different stages of optimization. To preserve the training behavior of the full dataset, we introduce a multi-order matching objective that jointly aligns the first-order mean and projected uncentered second-order moments of gradient trajectories. GLOBE further combines Group LASSO, Elastic Net regularization, and nonnegative budget constraints to induce group- and sample-level sparsity while stabilizing the weights of correlated trajectories. Finally, class-balanced Top-K selection maintains adequate category coverage under limited sampling budgets. Experiments across six benchmarks and five evaluation architectures demonstrate that GLOBE consistently outperforms existing coreset selection methods in downstream test accuracy, particularly at low retention ratios. These results highlight the effectiveness of combining dynamic gradient information, multi-order distribution matching, and structured sparsity for data-efficient learning.

AI 总结：针对长上下文大模型KV缓存的INT2量化瓶颈，提出OptR输出感知旋转方法，通过分解误差、学习正交修正等，提升QuaRot等性能并增强长上下文检索能力，且开销可忽略。

链接：https://arxiv.org/abs/2608.02691

机构：University of Southern California（南加州大学）； Seoul National University（首尔大学）； Inha University（仁荷大学）

作者：Vincent-Daniel Yun, Woosang Lim, Minsoo Cheong, Sunwoo Lee, Murali Annavaram, Sai Praneeth Karimireddy, Sungjoo Yoo

英文摘要：The key-value (KV) cache has become a major memory and bandwidth bottleneck in long-context large language model inference, making ultra-low-bit quantization increasingly important. However, existing rotation-based INT2 methods optimize cache statistics or proxy errors before the complete attention readout, even though the model is ultimately affected by the error propagated through attention and the output projection $W_O$. To address this mismatch, we propose \textit{OptR}, an output-aware rotation method that minimizes post-$W_O$ attention-output error. OptR decomposes the post-$W_O$ attention-output error into key- and value-induced terms and learns per-head orthogonal corrections through the full INT2 quantization and attention path. OptR further applies an attention-equivalent key reparameterization to reduce large channel-wise offsets without changing the softmax distribution. Across three models and five reasoning and coding benchmarks, OptR consistently improves both QuaRot and OSCAR and strengthens long-context retrieval, while preserving the paged KV-cache format with negligible inference overhead.

AI 总结：该研究提出PatTree，一种自动构建的多模态图基患者表示，在ADNI-1队列子集上实现阿尔茨海默病等三分类任务98.5%平衡准确率，可作为临床AI流程的可扩展基础。

链接：https://arxiv.org/abs/2608.02692

作者：Julia Gehrmann, Lars Quakulinski, Hamza Naseem, Oya Beyan

英文摘要： Access to holistic, multimodal data improves the performance of Artificial Intelligence (AI) in medical classification tasks compared to utilizing single modalities or data sources. However, the inherent heterogeneity and complexity of clinical real-world data pose significant challenges to structured data analysis and AI application. This heterogeneity includes missing values, multiple time points, diverse modalities, and inconsistent formats and semantics. Data harmonization prior to data integration tackles this challenge but remains resource-intensive and error-prone, limiting the scalability and reproducibility of holistic, AI-driven decision support on clinical real-world data. We therefore propose PatTree, a graph-based, holistic representation of patients that can be derived from real-world clinical data through the automated structuring of multimodal clinical data. PatTree enables early-stage data integration without relying on pre-standardized inputs. While representing heterogeneous clinical data within a unified knowledge graph, PatTree preserves the semantic relationships between data elements across modalities and data sources, facilitating interoperability and machine-interpretable data access. Using a subset of the ADNI-1 cohort (n = 763), we demonstrate that classification of patients is directly feasible on PatTree reaching state-of-the-art classification performance. In the three-class classification task distinguishing Alzheimer's disease, mild cognitive impairment, and cognitively normal individuals, we achieve a balanced accuracy of 98.5% and an F$_1$ score of 0.987 on the held-out test set. Our results show that assumption-free, automated structuring of multimodal medical data can serve as a scalable foundation for clinical AI pipelines bypassing tedious data preparation and standardization.

AI 总结：本文提出基于分布的归因可分性框架，用于评估归因方法的稳定性及对比其排名鲁棒性，为归因方法稳定性评估提供补充准则。

链接：https://arxiv.org/abs/2608.02697

机构：Barcelona Supercomputing Center（巴塞罗那超级计算中心）

作者：Eddie Conti, Álvaro Parafita, Axel Brando

英文摘要：Attribution methods (AMs) assign an importance score to each feature and are widely adopted to explain black-box models. However, most methods can produce variable attribution scores due to stochastic components in their definition. In this paper, we propose a distribution-based framework to capture the stability of attribution scores. In particular, our approach allows to understand the degree of separability in the ranked attribution vector and obtain the largest index for which a feature ranking remains reliable. We further extend this framework to compare AMs based on the robustness of their rankings across a dataset. Through experiments, we demonstrate how to apply our method to evaluate explainer stability. Overall, our approach provides a complementary criterion for evaluating the stability of AMs.

AI 总结：针对模拟内存计算的噪声问题，提出NANQ量化框架，实现混合精度非均匀量化，在2比特量化下提升视觉模型准确率、降低语言模型PPL，仅用少量等效比特即可获显著增益。

链接：https://arxiv.org/abs/2608.02700

作者：Yizhe Chen, Wenshuai Yao, Saiya Wang, Yuannuo Feng, Wenbo Qi, Kechao Tang, Ngai Wong, Wenyong Zhou, Wang Kang

英文摘要：Analog compute-in-memory (CIM) enables energy-efficient neural network inference, but device variation and read noise can severely degrade low-bit quantized models. Existing CIM-oriented quantization methods mainly minimize ideal quantization error, ignoring the hardware noise floor and thus causing inefficient precision allocation. We propose NANQ, a noise-aware mixed-precision non-uniform quantization framework for analog CIM. NANQ models magnitude-dependent weight noise from measured responses of an eFlash CIM array and converts the noise profile into an adaptive quantization density, assigning finer resolution to low-noise regions while avoiding ineffective precision in noise-dominated regions. It further assigns layer-wise bit-widths by identifying each layer's precision saturation point under hardware noise using a unified threshold. On-chip experiments on an eFlash CIM SoC show that, under 2-bit weight-magnitude quantization, NANQ improves vision-model accuracy by 8.05 percentage points and reduces language-model PPL by 54.7% on average over PowerQuant. Mixed-precision NANQ captures most of the gains obtainable from additional quantization resources with only 3.2-3.8 equivalent bits.

AI 总结：该研究探讨训练日志能否提升随机训练模型比较的精度，采用针对各模型的协变量调整方法，经视觉实验发现早期训练日志的简单调整可降低比较不确定性，但协变量选择的噪声会限制效果。

链接：https://arxiv.org/abs/2608.02705

作者：Wei-Jung Huang

英文摘要：Comparing stochastically trained models requires estimating both a performance difference and its uncertainty from repeated runs. We study whether training logs from those same runs can make such comparisons more precise. Because training-log covariates are produced during training rather than measured before it, we use arm-specific covariate adjustment: each model is adjusted only with statistics from its own runs, and the raw mean difference remains the reported effect. In a vision study spanning three architectures and three datasets, simple adjustments based on early training logs often reduce uncertainty in model comparisons. The main limitation is covariate selection. Broadly searching the log pool for the most correlated statistic often adds more noise than it removes, even when useful statistics exist in hindsight. Training logs therefore appear useful for more precise model comparisons, but only when the adjustment avoids large selection noise.

AI 总结：该研究针对消息传递神经网络的虚拟节点内存瓶颈，提出可寻址且保留基数的虚拟内存方案，通过交叉注意力槽和私有锚点实现，在多重性感知等任务上验证了其有效性。

链接：https://arxiv.org/abs/2608.02709

作者：Félix Marcoccia

英文摘要：Virtual nodes give message-passing neural networks a simple global communication route, but the standard node--VN--node pipeline compresses the graph into one homogeneous state and broadcasts it identically to every node. Building on the Two-Radius analysis of Mishayev et al., we ask how auxiliary virtual memory can relieve this finite-capacity bottleneck without self-attention. We identify two requirements. First, the global memory should be factorized into independently writable and readable states: this can be achieved using addressable cross-attention slots. Second, addressability alone does not preserve multiplicity, because softmax attention is invariant to uniform replication. Inserting each slot query as a private key/value anchor recovers the discarded normalization mass and yields, on bounded color domains, an injective multiset representation able to implement a 1-WL refinement. Experiments on multiplicity-aware Two-Radius, motif counting, and constrained link-set prediction support this addressable and cardinality-preserving virtual memory at (O(nMd)) arithmetic cost.

AI 总结：该研究提出NNLCI神经网络，通过局部校正粗细网格的期权定价解，以少量高保真数据训练，将细网格解的RMSE降低4-12倍，可高效用于高维期权定价，降低计算需求。

链接：https://arxiv.org/abs/2608.02778

机构：School of Mathematics, Georgia Institute of Technology（佐治亚理工学院数学学院）

作者：Harris Cobb, Wenbo Hao, Yingjie Liu

英文摘要：We present a novel application of Neural Networks with Local Converging Inputs (NNLCI) to improve the efficiency of existing numerical methods for pricing multi-asset options. The most concise input format for NNLCI has been introduced, offering substantial convenience and efficiency. NNLCI uses a neural network to locally correct solutions from a coarse mesh and a refined mesh (relative to the coarse one), requiring only a minimal amount of high-fidelity training data. We demonstrate this approach on cash-or-nothing options under the Black-Scholes equation in one, two, and three spatial dimensions, and on single-asset down-and-out barrier call options under the Heston stochastic-volatility model (whose pricing PDE is two-dimensional in the spot price $S$ and the instantaneous variance $v$). In each case, NNLCI reduces the root-mean-square error (RMSE) of the refined-mesh numerical solution by a factor of approximately 4-12 on test sets, even when the neural network is trained on only a small subset of parameter combinations. These results demonstrate that NNLCI significantly reduces computational requirements for high-dimensional problems in real-time options trading and risk management, offering low training costs and strong generalization ability.

AI 总结：本文提出评估盲区概念，指出测量故障会从AI训练到部署阶段无声传播，通过案例与50起真实事件验证，发现53%公开失效为无声，强调测量基础设施需覆盖全生命周期。

链接：https://arxiv.org/abs/2608.02786

作者：Priyanka Bajaj (Independent Researcher)

英文摘要：AI systems can fail silently. The failure propagates through training loops, evaluation pipelines, and production monitoring stacks until downstream harm makes it visible. This paper introduces evaluation blindness: a measurement function M exhibits evaluation blindness with respect to failure class F when it produces readings indistinguishable from a healthy state while the system is actually failing, with no auxiliary signal flagging the gap. The problem surfaces at two lifecycle stages the literature has treated separately. At training time, reward models are gamed, importance-sampling corrections are silently miscalculated, and benchmark contamination inflates fine-tuning evaluations, all while loss curves look healthy and gradient updates proceed normally. At deployment time, monitoring fails to catch six classes of production failure, including an Operational category that is 100% silent by structural definition. We provide a formal detectability predicate unifying both stages. Four training-time case studies trace concrete breakdowns, including a real implementation bug in TRL PR #6594 where gradients are corrupted as loss decreases normally. A six-class taxonomy validated against 50 real-world incidents from court documents and regulatory filings finds that 53% of verifiable public failures were silent. A failure budget framework ties acceptable failure rates to use-case risk class. The implication is direct: measurement infrastructure is a correctness concern across the full AI lifecycle, not just at evaluation time. Data, code, and taxonomy schema are at this https URL.

AI 总结：该研究以 Pythia 模型家族为例，揭示 Transformer 尺寸间转换的关键在于初始化，提出最小二乘补偿与方差保持缩放两个控制项，在低预算下可高效实现模型转换，优于从头训练与子克隆方法。

链接：https://arxiv.org/abs/2608.02829

作者：Ravi Satya Durga Prasad Yenugula

英文摘要：Model families train every size from scratch. Can a pretrained large model be converted into a smaller sibling? We characterize the 1.4B->410M conversion in the Pythia family end-to-end: (i) representations align strongly across sizes (ridge R^2=0.84) while parameters align weakly; (ii) dense weight projection is functionally destructive -- provably not an assembly artifact -- because basis mixing breaks rotary, per-head, GELU, and LayerNorm structure; (iii) after the best-fit linear operator, weight residuals are statistically indistinguishable from noise under shuffle controls; (iv) conversion value therefore lives in initialization. In matched-budget continued pre-training we decompose conversion into two independent levers -- least-squares compensation (function: best zero-shot) and variance-preserving rescale (dynamics: best endpoints). Compensation is a token-efficient, low-budget win rather than a universal one: at 30M tokens it beats the strongest subcloning variant on both a width-reduced pair (84.0 +/- 1.8 vs. 89.7 +/- 3.7, 3/3 seeds) and a held-out depth-reduced pair (109.3 vs. 117.9, 3/3 seeds), reaching a given quality with fewer tokens; at a 33x larger budget the two converge to parity (40.0 vs. 40.0), both far ahead of from-scratch, which transfer initialization always beats -- by up to 18x at low budget, the margin narrowing at convergence and at the largest scale. We further map the method's boundary: at ~5x the donor scale (6.9B->1.4B) stacking both levers over-corrects, which we trace to ill-conditioning of the compensation solve at large width, pointing to dimension-aware regularization as the fix. Code, checkpoints, and the frozen evaluation corpus are released.

AI 总结：本文提出适用于多种模型的事后概念漂移缓解方法NOMADD，通过参数外推提升模型在18个数据集基准上的性能，训练耗时远低于同类最先进方法。

链接：https://arxiv.org/abs/2608.02845

作者：Swapn Shah, Keith Burghardt

英文摘要： Tabular model performance degrades when feature distributions change over time or the relationship between features and outcome variables change over time, known as data drift and concept drift, respectively. These issues are challenging to mitigate in real time because labeled data may not be immediately available, or re-training a model could be impractical. While tools exist to reduce drift, they are typically bespoke to neural network architectures and adapt how models are trained. In this paper, we offer an alternative post-hoc method to reduce concept drift, which is applicable to a variety of models, from trees to neural networks to tabular foundation models. This new tool is especially useful when constraints, such as high model accuracy, bounded inference time, or model size requires users to choose between different models for their specific use-cases. Our algorithm fits the base model separately on each labeled training period, measures how its parameters evolve against a single anchor model pooled over all of those periods, compresses those changes with a low-rank factorization, and extrapolates each latent factor forward with a damped, regularized forecast. On the 18-dataset Drift-Resilient TabPFN benchmark, evaluated under that benchmark's own protocol and metric, the extrapolation improves every base family it is applied to, and achieves performance competitive with the state-of-the-art Drift-Resilient TabPFN with seconds of training. In contrast, Drift-Resilient TabPFN requires pre-training on millions of synthetic datasets over approximately 1,300 GPU-hours, and is orders of magnitude slower in inference (depending on the model). In the discussion, we explore the promise and challenges of extending this tool to other modalities.

AI 总结：针对基于知识的视觉问答中多模态检索的负样本处理问题，提出贝叶斯数据重加权框架，通过概率建模与随机EM优化，在三个检索器和七个基准上提升了检索准确率。

链接：https://arxiv.org/abs/2608.02907

机构：University at Buffalo（布法罗大学）； NEC Laboratories America（美国 NEC 实验室）； Adobe Research（奥多比研究院）； Iowa State University（爱荷华州立大学）； New York University（纽约大学）

作者：Jingchen Sun, Shaobo Han, Ruiyi Zhang, Naresh Kumar Devulapally, Ming Liu, Yitao Long, Vishnu Suresh Lokhande, Changyou Chen

英文摘要：Multimodal retrievers are essential for knowledge-based visual question answering, where they retrieve external evidence for image-question pairs. However, existing contrastive training methods typically treat all unmatched query-document pairs as equally informative negatives, which is problematic because many unmatched documents may still be semantically relevant or partially useful. We propose Bayesian Data Reweighting, a probabilistic framework that models query-document importance as latent variables and adaptively infers posterior weights to downweight likely false negatives. With closed-form posterior updates under conjugate priors and stochastic EM optimization, our method consistently improves retrieval accuracy across three retrievers and seven knowledge-based VQA benchmarks.

AI 总结：本研究提出基于髋部加速度计数据预训练的CNN-BiLSTM模型CHAP，可迁移至腕部数据实现坐/非坐分类，微调后性能优于从头训练的Transformer，为腕部传感器久坐行为检测提供了有效方案。

链接：https://arxiv.org/abs/2608.02946

机构：Halıcıoğlu Data Science Institute, University of California, San Diego（加州大学圣迭戈分校哈利乔格鲁数据科学学院）； Herbert Wertheim School of Public Health and Human Longevity Science, University of California, San Diego（加州大学圣迭戈分校赫伯特韦特海姆公共卫生与人类长寿科学学院）； Department of Computer Science and Engineering, University of California, San Diego（加州大学圣迭戈分校计算机科学与工程系）； Center for Children’s Healthy Lifestyles & Nutrition, Children’s Mercy Kansas City, University of Missouri-Kansas City（密苏里大学堪萨斯城分校儿童慈善医疗堪萨斯城分院儿童健康生活方式与营养中心）； Kaiser Permanente Washington Health Research Institute（凯撒永久华盛顿健康研究所）； Department of Kinesiology and Nutrition, University of Illinois Chicago（芝加哥伊利诺伊大学运动机能学与营养系）

作者：Yuliang Chen, Weiwei Shi, Jingjing Zou, Rong Zablocki, Animesh Kumar, Jordan A. Carlson, Sheri J. Hartman, Mikael Anne Greenwood-Hickman, Paul R. Hibbing, Marta Jankowska, Jay Yang, Arun Kumar, Loki Natarajan

英文摘要：Accurate detection of sedentary behavior is important for studying health risks related to prolonged sitting, but posture-based classification remains challenging with wearable sensors, especially at the wrist. We study whether a deep learning model trained on hip-worn accelerometer data can transfer to wrist-worn accelerometer data for sitting versus non-sitting classification. We use CHAP, a CNN-BiLSTM model originally developed for hip accelerometers, and evaluate its zero-shot performance on wrist data as well as its adaptation through finetuning with varying amounts of labeled wrist data. Experiments are conducted on the iWatch dataset with ground-truth posture labels derived from wearable cameras. The hip-trained model performs strongly on hip data without retraining, but accuracy drops on wrist data due to sensor placement shift. Finetuning CHAP provides consistent advantages over transformer models trained from scratch. These findings suggest that hip-based pretraining provides a useful starting point for wrist deployment, while highlighting the need for wrist-specific adaptation to handle higher signal variability.

AI 总结：ATFlash提出逐RoPE波长注意力窗口，修剪查询-键内积项，在保持长上下文任务性能的同时，可移植至FlashAttention-4等框架，在多款模型上实现LLM推理的计算与内存效率提升。

链接：https://arxiv.org/abs/2608.02947

作者：Shun-ichiro Hayashi, Daichi Mukunoki, Tetsuya Hoshino, Takahiro Katagiri

英文摘要：The attention score with rotary position embeddings (RoPE) decomposes exactly into a sum over its 2D-rotation frequency pairs, and each pair's wavelength limits how far it can discriminate position. Aligned with this structure, we propose the per-RoPE-wavelength distance window: it prunes the query--key inner-product terms beyond a wavelength-proportional distance. Unlike a sliding window, every key remains reachable, at least through the low-frequency pairs. The reduction rate is input-independent, with a closed form logarithmic in the sequence length $N$, in contrast to dynamic-sparse methods like MInference. Such token-level selection is orthogonal to our frequency-level pruning. The window can therefore be applied on top of those methods. On Qwen2.5-0.5B and Llama-3.2-3B, the window prunes 37--48\% of the query--key inner-product terms within each model's native context length. Relative to full attention, the top-1 match rate stays at 96--98\% and the mean output-distribution KL at the $10^{-3}$-nat level on LongBench-v2 contexts. We examine absolute scores on long-context benchmarks such as RULER, OpenAI-MRCR, LongCodeQA, and $\infty$Bench: they are broadly preserved. We implement the window as a slice of the query--key contraction axis, leaving the online-softmax recurrences untouched, and port it with minimal diffs into the released FlashAttention-4 prefill and FlashInfer decode. On RTX PRO 6000 with Llama, both ports outpace stock with gains growing with context length, up to $1.29\times$ at 128K. End to end on Qwen2.5-7B-1M, with 57\% of the inner-product terms pruned, the speedup reaches $1.31\times$ at a 1M-token context.

AI 总结：该研究发现大型语言模型的反向转向向量（ISVs）会产生与预期相反的转向效果，提出无需生成即可区分ISVs的方法，通过推理时干预改进转向流程，在多数实验中提升了效果。

链接：https://arxiv.org/abs/2608.02957

机构：Northeastern University（东北大学）

作者：Max Torop, Aria Masoomi, Jennifer Dy

英文摘要：Steering vectors (SVs) are widely used to influence the expression of concepts (e.g., truthfulness) in large language model outputs. A key assumption underpinning SVs is that they are linearly discriminative with respect to the concept: representations of texts that exhibit the concept are more aligned with the SV than those that do not, motivating shifts along the positive or negative SV direction to respectively promote or suppress the concept. In this work, we identify an inverted detection-control phenomenon in which some highly discriminative SVs that are aligned with positive representations can consistently promote the opposite behavior. We refer to such vectors as inverted-steering vectors (ISVs). We provide a geometric characterization of ISVs' effects, finding that steering along these directions systematically pushes representations in discriminative downstream heads as if the concept were absent, even prior to decoding. Motivated by this analysis, we propose an approach for distinguishing ISVs without requiring generation or associated response scoring. This enables targeted sign flips, which we use to improve a foundational detection-based steering pipeline via Inference Time Intervention (ITI). Our approach improves results in 27/30 experiments, ranging from +0.9% to +138%. We evaluate our findings on Gemma 3 12B, Qwen 2.5 14B, and Olmo 3 7B across 5 concepts.

AI 总结：该研究指出LLM回测的标准污染检测方法无效，提出利用已知截止时间和匹配干净对照组测量时间泄漏的方法，可检测并调整回测得分，澄清部分模型优势源于近期性而非真实技能。

链接：https://arxiv.org/abs/2608.02985

机构：Northwestern University（西北大学）

作者：Zeyu Zhang, Bradly C. Stadie

英文摘要：The standard check for contamination in LLM backtests is simple: compare scores before and after the training cutoff. We show this check is uninformative. Four flagship models fail it on questions they cannot have memorized: every scored question resolved after their cutoffs. The reason is structural. Models legitimately know more about times near their cutoff, so recency mimics leakage, and we prove no passive backtest can separate the two from genuine skill. Measurement, not just detection, requires information from outside the backtest. We supply it in two forms. A known cutoff identifies leakage at the boundary; a matched clean control identifies it globally and yields a leakage-adjusted score. We also derive where leakage hides: it concentrates on outcomes that surprised the crowd and were well covered in training, and partial memorization is disproportionately rewarded. We validate the estimators against ground truth by planting leakage in twin models, where they recover the injected dose and return null on clean questions. Deployed on frontier models, they detect one cutoff-localized signature and, at the audit's power floor, clear five models whose apparent advantages were recency alone. Backtests need not be discarded; they need one defensible reference.

AI 总结：该研究针对IMDb数据集训练BERT-mini，对比不同谱优化方法，提出的联合正则化逆方法可提升测试准确率、降低损失，是仅权重谱优化的小幅但一致的扩展。

链接：https://arxiv.org/abs/2608.02991

机构：Harbin Institute of Technology, Shenzhen（哈尔滨工业大学（深圳））

作者：Gongyue Zhang, Honghai Liu

英文摘要：Matrix spectral optimizers reshape weight-update spectra but usually delegate vector-valued biases to a separate optimizer. We study whether this separation is neutral. We formulate each affine layer as a joint momentum matrix $A=[M_W,\alpha m_b]$ and apply a capped regularized-inverse spectral map to the complete matrix, producing both the weight and physical bias updates. A strict five-seed ablation on a four-layer BERT-mini trained from scratch on IMDb compares exact-SVD Muon, weight-only inverse shaping, affine-probe inverse shaping, and the proposed joint regularized inverse (JRI). Weight-only inverse shaping raises validation-loss-selected test accuracy from $84.903\pm0.242\%$ to $85.562\pm0.308\%$ and lowers selected test loss from $0.3479$ to $0.3345$. Allowing bias to alter the joint SVD while retaining an independent Adam bias update does not improve over weight-only inverse shaping. Using the transformed bias jointly raises selected test accuracy to $85.738\pm0.180\%$ and lowers test loss to $0.3291$, with all five seeds improving relative to the probe baseline. During the peak-performance window, JRI preserves the eligible weight-update norm while reducing the bias-update norm from $0.02095$ to $0.00301$, lowers boundary-function share from $86.58\%$ to $78.97\%$, and changes the cosine between weight-induced boundary motion and explicit bias from $+0.030$ to $-0.137$. An independent 22-seed replication yields $85.743\pm0.203\%$ selected test accuracy. These results identify joint affine spectral allocation as a small but consistent extension to weight-only spectral optimization.

AI 总结：本文从图信号处理视角研究LLM上下文学习中数值序列的表示，发现数值推理的内部特征随上下文长度和输入动态复杂性变化，且在不同模型家族间一致。

链接：https://arxiv.org/abs/2608.03015

机构：Cornell University（康奈尔大学）； Goodfire AI（古德火人工智能公司）； Imperial College London（伦敦帝国学院）

作者：Jiajun Bao, Zihao Qi, Toni J.B. Liu, Gurbir Arora, Raphaël Sarfati, Nicolas Boullé, Christopher J. Earls

英文摘要： Pretrained large language models (LLMs) have demonstrated in-context learning (ICL) capabilities for numerical inference over sequences serialized as text. Prior work has identified and characterized this form of numerical inference primarily through output-level evaluations such as prediction error. However, how numerical information is organized within LLM representations remains much less understood. To study this internal organization, we adopt a graph signal processing perspective in which attention induces a weighted graph over tokens, while token hidden states define signals on its nodes. Quantitative graph-spectral diagnostics and qualitative token-graph visualizations reveal that representations become more clearly differentiated by input dynamical complexity as context length increases. Simpler inputs produce attention-induced token graphs with stronger global connectivity and smoother, spectrally concentrated hidden-state signals, whereas more complex inputs produce more localized graphs and hidden-state signals with broader spectral support and greater high-frequency energy. Together, these findings point to systematic, context-dependent internal signatures associated with numerical ICL that are conserved across model families.

AI 总结：本研究针对 deceased donor 肾移植的存活预测，提出基于配对受者的评估框架，发现五类模型均达约60%准确率，且该指标比C-index更具临床相关性。

链接：https://arxiv.org/abs/2608.03017

机构：Case Western Reserve University（凯斯西储大学）； University of Toledo（托莱多大学）

作者：Misaki Matsuura, Mohammadreza Nemati, Dulat Bekbolsynov, Stanislaw Stepkowski, Kevin S. Xu

英文摘要：There has been significant interest in using machine learning algorithms to predict kidney transplant outcomes, such as the number of years until a graft inevitably fails. These prediction algorithms could possibly be used for pre-transplant donor-recipient matching to identify more compatible donors and recipients and thus improve post-transplant outcomes. In this study, we explore the use of survival prediction models trained on deceased donor kidney transplant data from the Scientific Registry of Transplant Recipients (SRTR). We propose a novel paired recipient-based evaluation framework that compares graft outcomes between two recipients who received kidneys from the same deceased donor, allowing us to evaluate the counterfactual benefit of changing the recipient for a certain donor. We find that five different survival prediction models, ranging in complexity from linear to deep learning-based models, all result in ~60% paired recipient-based accuracy. We further translate this accuracy into an interpretable quantity of post-transplant years gained. We also highlight major limitations of the commonly used concordance index (C-index) metric for evaluating survival prediction accuracy in this setting and demonstrate that our proposed paired recipient-based accuracy metric is more clinically relevant and better reflects real-world allocation settings.

AI 总结：本文提出SynEnergy，一种两阶段扩散框架，通过HG-ASL提取异常语义、AS-Diff生成数据，在四个真实能源数据集上较11种基线提升异常保留与下游质量，可扩展至城市级场景。

链接：https://arxiv.org/abs/2608.03087

作者：Lin Jiang, Dahai Yu, Ravikumar Gelli, Guang Wang

英文摘要：Fine-grained energy consumption data are essential for applications such as demand forecasting, demand response planning, and grid reliability assessment. However, access to such data is often restricted by privacy concerns and data-sharing constraints, motivating growing interest in synthetic energy data generation. Although existing methods can reproduce overall consumption distributions and recurring temporal patterns, they often smooth out or underrepresent anomalous events caused by extreme weather, infrastructure failures, and behavioral shifts. Preserving these events is challenging because they are sparse, localized in time and space, and shaped by heterogeneous dependencies across geographical proximity and regional attributes. To address these challenges, we propose SynEnergy, a two-stage diffusion-based framework for anomaly-preserving energy consumption data generation. The first stage, Heterogeneous Graph-based Anomaly Semantic Learning (HG-ASL), extracts region-specific anomaly semantics from sparse residual structures by jointly modeling spatial and attribute dependencies across urban regions. The second stage, Anomaly Semantic-guided Diffusion (AS-Diff), injects the learned anomaly semantics into the denoising process to generate realistic consumption sequences while preserving anomalous patterns. This design enables controllable generation for individual regions and scales naturally to city-wide settings. We evaluate SynEnergy on four real-world energy consumption datasets against 11 general-purpose and energy-specific generation baselines. Experimental results show that SynEnergy improves anomaly preservation fidelity by an average of 12.21% and downstream quality by 2.96%, while maintaining competitive overall generation fidelity compared to baselines.

AI 总结：该研究提出分裂候选数量作为梯度提升决策树（GBDTs）的单轴容量参数，通过分析其与双重下降现象的关联，经XGBoost等模型实验验证了分裂候选预算对测试误差的影响规律。

链接：https://arxiv.org/abs/2608.03111

作者：Ryuichi Kanoh

英文摘要：Double descent is commonly studied by scaling an explicit capacity parameter, such as neural-network width. For gradient boosting decision trees (GBDTs), however, an analogous single-axis capacity parameter has not been established. We propose the number of split candidates as an operational capacity parameter for GBDTs. Holding other training controls fixed, increasing the split-candidate budget refines the feature-quantization grid and expands the dictionary of root-to-leaf paths from which boosting selects its updates. To analyze this expansion, we construct an empirical tree-kernel diagnostic that summarizes how candidate-induced paths group the training examples. A regime in which the empirical kernel rank grows toward the sample size and very small positive eigenvalues emerge exposes noise-sensitive directions; in this regime, test error peaks before decreasing again at larger split-candidate budgets. This perspective predicts that deeper trees should reach the regime with fewer split candidates, larger training sets should require finer grids, and label noise should make the peak more pronounced. Experiments support these predictions and show test-error peaks at intermediate split-candidate budgets across XGBoost, LightGBM, and CatBoost, whereas a random-forest control improves monotonically under the same split-candidate sweep. Taken together, our analysis and experiments support split-candidate scaling as a single-axis capacity intervention for studying GBDTs and suggest that the observed double descent arises from an interaction between candidate-induced geometry and boosting dynamics.

AI 总结： 该研究针对SAM偏向平坦极小值的隐式偏差，通过线性稳定性分析得到定量超参数边界，验证了增大ρ可降低Hessian最大特征值，并提出TLC-SAM变体进一步优化性能。

链接：https://arxiv.org/abs/2608.03197

作者：Jiaxin Deng, Junbiao Pang

英文摘要：Sharpness-Aware Minimization (SAM) improves generalization by seeking parameters whose loss is robust to local adversarial perturbations, but the quantitative mechanism underlying its implicit bias toward flat minima remains unclear. In particular, the perturbation radius $\rho$ is typically treated as an isolated tuning parameter, despite defining the neighborhood in which SAM measures sharpness. We analyze mini-batch SAM near an interpolating minimum through linear stability. Under local linearization and gradient-noise alignment assumptions, we prove that every linearly stable minimum satisfies $\lambda_{\max}\leq\sqrt[3]{b\Gamma/(2\rho\eta^2)}$, where $\lambda_{\max}$ is the largest Hessian eigenvalue, $b$ is the batch size, $\eta$ is the learning rate, and $\Gamma$ bounds the gradient norm. The bound quantitatively characterizes SAM's implicit flatness bias: holding the other quantities fixed, a smaller batch size, a larger learning rate, or a larger radius restricts linearly stable SAM to flatter minima. It also exposes a necessary trade-off: $\rho$ should be large enough to promote flatness, yet remain local enough to preserve the approximation and stable training. We validate this prediction in a controlled study of 900 models on CIFAR-100 with ResNet-18 and VGG-19, where increasing $\rho$ is consistently associated with a smaller largest Hessian eigenvalue across batch-size and learning-rate settings. Finally, we instantiate the analysis in Taylor-Locality Controlled SAM (TLC-SAM), which adjusts $\rho$ using the observed Taylor-approximation error and further reduces the top Hessian eigenvalue relative to fixed-radius SAM. Our results provide quantitative hyperparameter bounds and a stability--locality perspective for analyzing and designing SAM variants.

AI 总结：本研究提出ED-DiT，一种物理引导的扩散Transformer，通过电子密度点云自监督预训练学习可迁移分子表示，在6项EDBench任务及低监督场景下均优于基线，展现出良好效果。

链接：https://arxiv.org/abs/2608.03260

作者：Liang Shuang, Haocheng Wang, Jiayi Song, Shuquan Ye, Ben Fei

英文摘要：Pretraining has shown strong potential for learning transferable representations, yet it remains underexplored for electron-density-based molecular learning. Electron density provides a continuous three-dimensional description of molecular electronic structure, capturing both local spatial patterns and global physical quantities. This raises a key question: can electron-density fields be used for self-supervised pretraining to learn a shared representation that transfers across diverse electronic-structure-related tasks? We propose ED-DiT, a physics-guided Diffusion Transformer for self-supervised pretraining on electron-density point clouds. ED-DiT learns reusable representations by reconstructing corrupted and partially masked log-density fields across diffusion noise levels. An electron-number consistency constraint is further introduced to preserve the total electronic mass. The pretrained encoder can be adapted to property prediction, open-/closed-shell classification, molecule-electron-density retrieval, and molecule-conditioned electron-density prediction. Experiments on six EDBench tasks show that ED-DiT consistently outperforms the same architecture trained from scratch, especially under limited supervision. For molecule-conditioned electron-density prediction, it reduces RMSE from 2.2474 to 1.3753 and surpasses the available baseline. With only 10% labels, it improves orbital energy prediction RMSE from 0.0293 to 0.0138. These results demonstrate the effectiveness of physics-guided electron-density pretraining for learning transferable molecular representations.

AI 总结：本研究验证潜在推理模型的“组成激活”为真实计算，复现30M参数循环深度推理器，发现激活存在于读出层，决策时边际跃升，隐藏状态方向突变后冻结，撤回早期速度低谷声明。

链接：https://arxiv.org/abs/2608.03263

作者：Simon Lam-Muir

英文摘要：We test whether the "compositional ignition" reported in latent-reasoning models is real computation, an instrument artifact, or inherited from verbal training data. We grow an independent realization of a published 30M-parameter recurrent-depth reasoner from scratch (same recipe and seed), film its development, certify fidelity through a pre-registered whole-signature gate, and measure resolution in two channels at once: the vocabulary readout and the hidden state. The ignition is real and lives at the readout: arrival time rises lawfully with problem depth, resolution is sharp and holds, and the signature reproduces across two same-seed realizations with divergent training trajectories. At commitment the decision margin jumps 5.8-8.0 logits in one iteration, exceeding the 90th percentile of near-threshold non-event steps in 96% of cases; the signed margin's zero-crossing there is definitional and carries no evidential weight, so the evidence is that conditioned magnitude. The hidden-state direction snaps in raw geometry, meeting its pre-registered criterion (in the decoder's LayerNorm coordinates it attenuates just below our bar, so the composite decoder-coordinate claim is not confirmed), and then freezes in both (descriptively so in decoder coordinates; angular steps 52.9 to 1.2 degrees over eight iterations), while subsequent displacement is predominantly radial (0.961 of squared-norm) and readout-null to a measured bound (radial logit effect <=5.7e-6). An earlier velocity-trough claim is withdrawn: pre-registered normalization controls showed it coordinate-dependent. Intermediates were never recoverable through the tied readout (relay 0.00). All criteria were frozen before their data; the predictions ledger, including this paper's own withdrawn headline, ships in the companion repository.

AI 总结：本研究利用思维链动态特性，在不假设语言化CoT语义忠实性的情况下，检测大语言模型在布尔可满足性任务中的分布式推理失败，并通过针对性提示干预提升了Llama3-70B的准确率。

链接：https://arxiv.org/abs/2608.03291

作者：Shashwat Sourav, Aishwarya Balwani

英文摘要： Chain-of-thought (CoT) reasoning improves large language model (LLM) performance while also providing an observable interface to the model's reasoning process. Existing approaches that leverage verbalized CoTs to monitor reasoning correctness, however, largely evaluate the semantic correctness or consistency of individual intermediate steps, rather than how the reasoning process evolves across the trace. As a result, failures distributed across the reasoning trajectory, rather than those localized to a single incorrect step, remain comparatively underexplored. Furthermore, verbalized CoTs need not faithfully reflect the model's internal reasoning, motivating analyses that do not treat individual statements as literal accounts of internal computation. In this work, we therefore ask whether the dynamics of visible CoT can be leveraged to systematically distinguish successful from failed reasoning without assuming such semantic faithfulness. We study a range of LLMs on verifiable Boolean satisfiability tasks with variable complexity, enabling controlled comparisons near each model's capability frontier. Tagging CoT sentences by reasoning function reveals premature verification collapse on SAT problems: incorrect traces enter clause checking earlier, repeat similar operations, and finalize sooner. On UNSAT problems, models presumptuously move towards incorrect SAT conclusions, checking candidate assignments rather than deriving contradictions across constructed cases. Subsequently, a targeted proof-search prompt intervention raises Llama3-70B accuracy from 13.3% to 85%, correcting 84.6% of these errors. These results show that capability failures can manifest as distributed, task-dependent changes in the structure of visible reasoning, and that CoT dynamics agnostic to whether the verbalized trace reflects the model's internal computations can help diagnose and correct failures.

AI 总结：该研究提出无需子空间假设的多头注意力学习算法，通过特定查询设计恢复规范头，可精确恢复参数且适用于近似输出，还扩展至单层 Transformer 的学习。

链接：https://arxiv.org/abs/2608.03294

作者：Sunyeop Kim, Insung Kim, Jian Guo

英文摘要：We study the problem of learning multi-head softmax attention from black-box input-output access. The learner may query arbitrary real-valued token sequences and observe only the scalar output at the final token. Recent work gives an algorithm using $O(d^2)$ value queries to recover the single-head parameters $(W,v)$. For multiple heads, the same work establishes identifiability under the assumption that the heads occupy pairwise orthogonal subspaces. Applying the single-head recovery algorithm separately to the heads additionally requires bases for these subspaces to be known. We recover a canonical representation by merging heads with the same $W_h$, summing their corresponding $v_h$, and discarding a merged head when this sum is zero, without these subspace assumptions. By varying the number of copies of a token, our algorithm obtains samples of a rational function whose interpolation separates the canonical heads. Additional queries formed by adding selected token vectors then match the same head across different queries. When the oracle outputs and all subsequent computations are exact, the learner chooses its query vectors at random and recovers the canonical pairs $\{(W_h,v_h):h\in[H]\}$ up to permutation with probability one. When $H$ is known, it uses exactly $4Hd^2-2H+1$ value queries of maximum length $2H+1$. If only a known upper bound $H_0$ is available, the algorithm uses $4H_0d^2-2H_0+1$ value queries of maximum length $2H_0+1$. For approximate oracle outputs, we give conditions under which the parameter error is at most a model- and query-dependent constant multiple of the output error. Finally, we extend our result to a one-layer Transformer with multi-head attention followed by a bias-free ReLU feed-forward network. Under additional conditions, we recover a functionally equivalent Transformer without relying on a separate algorithm for learning the feed-forward network.

AI 总结：该研究针对 ReLU NTK 格拉姆矩阵的最小特征值，推导了与维度无关的下界，构造匹配上界证明其速率紧，明确了该特征值的最坏情况边界。

链接：https://arxiv.org/abs/2608.03368

作者：Zhao Song

英文摘要：For $n$ unit vectors $x_1,\ldots,x_n \in \mathbb{R}^d$, we study the continuous ReLU derivative Gram matrix $H$, whose entries are obtained by averaging pairwise gated inner products over a standard Gaussian direction. Writing $ \Delta_\pm:= \min_{i \neq j} \min\{ \|x_i-x_j\|_2, \|x_i+x_j\|_2 \} $ for their projective separation, we prove the universal dimension-free lower bound $ \lambda_{\min}(H) = \Omega( \Delta_\pm/\sqrt{\log n} ) $. Conversely, we construct worst-case families satisfying the matching upper bound $ \lambda_{\min}(H) = O( \Delta_\pm/\sqrt{\log n} ) $, showing that this rate is tight up to universal constants.

AI 总结：本研究评估大型语言模型的推理接口，测试提示词和训练设置对推理长度、准确率的影响，发现早答指令和低 effort 推理在特定 token 限制下可提升部分数据集准确率，需综合报告多维度评估指标。

链接：https://arxiv.org/abs/2608.03401

作者：Francesca Carlon, Vincent Ginis, Andres Algaba

英文摘要：Large language models often reason at length before answering, increasing cost and latency. Prompts and trained settings can shorten this reasoning, but a shorter trace may only show that the model stopped sooner. Here, we evaluate paired runs of the same question at matched reasoning horizons across 198 GPQA Diamond and 500 MMLU-Pro questions. We test a numeric/concision prompt that announces a token limit for Qwen3-14B and the trained effort settings of gpt-oss-20b and -120b. The Qwen prompt shortens reasoning traces by 12-17%, while accuracy changes at matched token limits are small and mixed. A concise/early-answer instruction raises MMLU-Pro accuracy by 3.8 percentage points at 512 tokens, including +2.7 points when both runs are unfinished. Its gain at 2,048 tokens is uncertain. For gpt-oss, candidate-logit answers from completed low- and medium-effort reasoning are 14.5-26.3 points more accurate than matched-horizon high-effort answers. Most of the 512-token advantage comes from lower effort finishing earlier, while differences among unfinished runs are smaller and mixed. Wrong early answers often concentrate probability on the chosen option, so earlier stopping does not uniformly improve probability quality. In these tests, a tight deadline can favor lower effort or a concise instruction, whereas allowing high effort to finish can recover higher final accuracy. Evaluations should report correct completion before a deadline, the answer obtained when a run is stopped, differences among unfinished runs, and probability assigned to the correct answer separately.

AI 总结：针对标签噪声学习中重构方法的隐式耦合问题，提出TRACE双源可靠性评估框架，分别评估观测标签与伪目标，提升了带噪声基准上的重构基线性能与伪监督可靠性。

链接：https://arxiv.org/abs/2608.03432

作者：Wenxiao Fan, Kan Li

英文摘要：Refurbishment-based noisy-label learning mixes an observed label with a model-derived pseudo target, typically using one sample-wise cleanliness score to control both branches. This creates a hidden coupling: reducing trust in the observed label automatically increases trust in the pseudo target. We show that this complementarity can replace one unreliable signal with another because a pseudo target learned from corrupted supervision may reproduce the noise it is meant to correct. Our representation diagnostics provide a consistent account of this mismatch: noisy supervision redirects deeper layers more strongly, whereas shallower relations remain comparatively stable and provide information beyond the loss posterior. We therefore propose TRACE, a Two-Source Reliability Assessment framework for Label Correction and Sample Reweighting. TRACE assesses the observed label using loss fit, shallow relation stability, and prediction agreement, while separately assessing the pseudo target using model confidence. Its source-specific scores control target correction and supervision strength without assuming complementary reliability. Across synthetic and real-world noisy benchmarks, TRACE improves representative refurbishment baselines and yields more reliable pseudo supervision.

AI 总结：本文提出无需训练的Approximate Speculative Decoding（ASD），通过带预算的最长前缀选择优化投机解码，提升了生成吞吐量与验证器接受率，且无需新草稿模型或微调。

链接：https://arxiv.org/abs/2608.03447

作者：Yuannuo Feng, Zegang Peng, Yuxin Xie, Yubing Ye, Yizhe Chen, Wenshuai Yao, Wenyong Zhou, Wang Kang

英文摘要：Speculative decoding accelerates autoregressive generation by verifying a draft block with a target model in parallel. Under standard greedy verification, decoding stops at the first draft token that differs from the target argmax, discarding the remaining target-scored suffix. Although accepting such a mismatch changes the decoding trajectory, it can make a contiguous suffix reusable when its tokens remain target-greedy under the realized prefix. In this paper, we introduce \textbf{Approximate Speculative Decoding (ASD)}, a training-free verifier that replaces binary first-mismatch truncation with budgeted longest-prefix selection. ASD accepts selected mismatches subject to a local target-logit regret gate, a per-block exception cap, and a persistent request-level regret budget, then reuses the contiguous target-greedy suffix without additional approximate decisions or target-model forward passes. ASD requires neither a new draft model nor fine-tuning, and exactly reduces to standard greedy verification when the budget is zero. Experiments show that ASD improves fixed-workload throughput by $3.05\%$--$15.26\%$ over matched strict verification and averages a $7.78\%$ gain across seven Qwen3-14B + DSpark-14B tasks. On DeepSeek-V4-Flash (284B) with DSpark it also raises verifier-side acceptance by roughly $10\%$--$16\%$ on GSM8K and MATH-500 in an FP4-to-FP8 compatibility setting. The source code is publicly available at: this https URL

AI 总结：该研究提出ConformalShift攻击，通过重排序真实前置事件降低心电监测的心室阈值，在MIT-BIH和INCART数据集上成功抑制心室类别，表明医疗自适应监测器可被真实信息时序破坏。

链接：https://arxiv.org/abs/2608.03628

作者：Arash Vashagh, Yasmin Vashagh

英文摘要：Adaptive conformal prediction can recover clinically important heartbeat classes missed by a point classifier, but delayed feedback makes its decisions sensitive to event order. We introduce ConformalShift, a bounded event-reordering attack that suppresses the ventricular class for rescued events without modifying ECG waveforms, labels, classifier scores, or the event multiset. ConformalShift searches for feasible permutations of authentic preceding events that lower the ventricular threshold before a selected target is evaluated. On disjoint MIT--BIH confirmation records, the attack suppressed 66.7% of eligible targets for Extra Trees and 60.0% for HistGradientBoosting, compared with random-schedule rates of 4.4% and 12.0%, respectively. Transferred configurations also outperformed random scheduling on INCART, while reducing the displacement budget weakened the attack on both datasets. These results show that adaptive monitors in healthcare can be compromised through the timing of authentic information, even when waveforms, labels, classifier outputs, and event contents remain unchanged.

AI 总结：该研究提出LAEF，一种7M参数的导联无关心电图基础模型，预训练于9.2M份12导联心电图，在18个下游数据集的1-2导联即时诊断场景下，性能优于零填充替代方案，与更大规模的12导联基线相当。

链接：https://arxiv.org/abs/2608.03690

作者：Edoardo Coppola, Stefano Fiorini, Pietro Liò, Mattia Savardi, Alberto Signoroni

英文摘要：Point-of-care cardiac devices such as smartwatches and handheld ECG recorders typically capture 1--2 leads, yet existing ECG foundation models are architecturally constrained to fixed 12-lead inputs, degrading or failing under these reduced configurations. We introduce LAEF (Lead-Agnostic ECG Foundation), a 7M-parameter ECG foundation model that can natively process any lead subset without zero-padding or architectural modification. LAEF represents ECGs as variable-size spatiotemporal graphs with physiologically motivated intra- and inter-lead connectivity, processed by a Graph Attention Network that scales naturally with active lead this http URL -trained on 9.2M 12-lead ECGs via masked node modelling with stochastic lead sampling, LAEF learns representations robust to lead configuration. Across 18 downstream datasets, LAEF is on par with specialized 12-lead baselines over 12$\times$ larger at full lead availability. Under direct point-of-care-oriented diagnostics (1--2 leads), it outperforms all zero-padded alternatives on 17 out of 18 datasets with with a single randomly sampled lead and on 14 out of 18 with 2 leads, with an average AUROC gain of +3.2 points. Representation analysis links this advantage to architectural lead-agnosticism, and a lead-importance study across 164 cardiovascular conditions shows population-level performance is stable across single standard input leads while still recovering established clinically lead-condition associations.

AI 总结：本报告面向非专业人士总结经典统计学习模型与算法，从范畴论视角解读统计学习模型，以吸引其他领域研究者参与该领域研究。

链接：https://arxiv.org/abs/2608.03706

作者：Congwei Song

英文摘要：Statistical learning is a fascinating field that has long been the mainstream of machine learning/artificial intelligence. A large number of results have been produced which can be widely applied to real-world problems. It also leads to many research topics and also stimulates new research. This report summarizes some classical statistical learning models and well-known algorithms, especially for amateurs, and provides a category-theoretic perspective on understanding statistical learning models. The aim is to attract researchers from other fields, including basic mathematics, to participate in the research related to statistical learning.

AI 总结：该研究提出了RESUME CONTRACT契约，用TLA+模型等方法验证了五个智能体工作流框架的一致性问题，并通过REMIT修复了相关故障。

链接：https://arxiv.org/abs/2608.03836

作者：Sajjad Khan

英文摘要：A framework that persists execution state so a run can be interrupted, survive a crash, and continue must decide what a resume means for effects that already fired. Five widely deployed agent workflow frameworks answer differently, none exposes a machine-checkable contract, and behavior violates even the fragments they state. The RESUME CONTRACT states six properties over the persistence API (prefix continuation, effect exactly-once, fork determinism, checkpoint validity, consume-once, recovery determinism), plus fork-intent and liveness obligations. A TLA+ model checks a reference semantics exhaustively, unchanged at scaled bounds (7.4 million states); a 39-cell fault matrix yields the separating models independence requires, and consume-once splits, its consumption clause independent of all six others. A deterministic, LLM-free harness measures them at pinned releases. LangGraph 1.2.9 durably records a second resume value and never consults it, persists schema-invalid state silently, and re-executes durably recorded work after a real SIGKILL: exactly-once across interrupts, at-least-once across crashes, on one API. CrewAI 1.15.2 re-executes completed effect-bearing methods against its written claim; pydantic-graph 1.x cannot resume after a mid-node crash; no two probed frameworks share a conformance profile. Consume-once holds sequentially and fails under concurrent delivery: k processes resuming one parked interrupt fire the gated effect k times, saturation 1.0 in 36 of 40 cells, and the failure crosses hosts. REMIT, a reference sequencer whose Verus-verified recovery core is line-identical to the shipped executable, repairs the fork and validity cells. The cross-process cell is repaired at the read path, and that repair ships: an opt-in gate claims consumption in the shared store, serving one racer and refusing the rest before any node executes.

AI 总结：针对不完整临床EHR数据的急诊分诊挑战，本文提出CRS-Triage模型，通过分模态评估可靠性并结合一致性估计置信度，权衡风险与覆盖，在MIMIC-IV-ED数据集上表现优异。

链接：https://arxiv.org/abs/2608.03862

作者：Guan Qiang, Yushen Chen, Tianlong Liu, David Rotenberg, Ethan H.Kim, Fang Fang

英文摘要：Emergency triage requires reliable decisions within a short time period. However, the available electronic health record (EHR) data, including structured data and clinical text, are often incomplete, unreliable, and inconsistent. This makes machine learning (ML)-based triage prediction more challenging, as existing ML models typically rely on complete and reliable EHR data to accurately predict patients' acuity levels. To address this, we propose confidence- and reliability-aware selective triage (CRS-Triage) to predict patients' acuity levels with a confidence score. By comparing the confidence score with a predefined threshold, CRS-Triage can selectively determine whether the model should make the decision or defer the case. Specifically, CRS-Triage separately evaluates the reliability of structured data and clinical text and then jointly considers the consistency between the two modalities to estimate the confidence of each prediction. Moreover, to reduce the risk of missing high-acuity patients, namely under-triage, CRS-Triage prefers to assign patients slightly higher acuity levels, namely over-triage, by penalizing under-triage errors. Experiments on the MIMIC-IV-ED dataset show that CRS-Triage achieves strong predictive performance. It also provides a better risk-coverage trade-off and remains reliable when the available EHR data are incomplete, degraded, or inconsistent across modalities.

AI 总结：针对轨迹推断现有算法计算成本高的问题，提出加速度匹配（AM）算法，通过提升至相空间回归条件加速度场，在基准问题上表现优于或相当现有算法。

链接：https://arxiv.org/abs/2608.03916

机构：University of Padova（帕多瓦大学）； École polytechnique（巴黎综合理工学院）； Yale University（耶鲁大学）

作者：Bartolo Dazzini, Giovanni Conforti, Alain Durmus, Aram-Alexandre Pooladian

英文摘要： Trajectory inference is a fundamental problem in many scientific domains: given a collection of unpaired snapshots of observations at discrete time points, the goal is to generate smooth trajectories that best resemble and interpolate the data. Existing algorithms exhibit computational challenges: they either rely on preprocessing subroutines to enforce smoothness or on simulation-based training objectives, both of which can be expensive. In order to overcome these limitations, we propose a new algorithm called Acceleration Matching (\texttt{AM}). Our approach consists of lifting the original interpolation problem to phase space and then regressing onto an explicit conditional acceleration field that induces random, smooth trajectories that agree with the prescribed marginals. Importantly, our resulting training algorithm only requires positional data, avoids trajectory simulation during training, and is devoid of expensive preprocessing. We provide ample numerical evidence suggesting that \texttt{AM} is competitive with or superior to existing algorithms on several benchmark problems from the existing literature.

AI 总结：该研究提出物理启发式神经网络（PFNN），将拉伸指数物理模型集成到CNN-Transformer中，通过混合训练范式实现工程化骨骼肌组织收缩动力学的高保真参数化，为高通量生物物理研究提供可靠工具。

链接：https://arxiv.org/abs/2608.03927

作者：Mattias Luber, Timo Betz

英文摘要：Engineered Skeletal Muscle Tissues (ESMs) have become a key structure for biomedical disease modeling and pharmacological screening, yet their functional characterization often relies on simplistic metrics like peak force, discarding critical kinetic information. This is partially due to the high level of mathematical complexity which mechanistic models introduce to capture these dynamics. Hence, exactly the complexity prevents scalable application and widespread adaptation in the field. Here we present a Physics-Flavored Neural Network (PFNN) that automates the kinetic phenotyping of ESMs. Our architecture integrates a stretched-exponential physical model into a CNN-Transformer, enabling the extraction of physically meaningful parameters directly from force-time profiles. To address the scarcity of labeled biological data, we employ a hybrid training paradigm: the model develops a "physical intuition" on synthetic data before undergoing unsupervised self-alignment on unlabeled real-world measurements. Our results demonstrate that this physics-flavored approach achieves high-fidelity parameterization across diverse contractile phenotypes and cell lines, including Duchenne Muscular Dystrophy models. Our scalable, self-improving pipeline bridges the gap between idealized biophysics and noisy \emph{in vitro} data, providing a robust tool for high-throughput biophysical research.

AI 总结：该研究针对推理大语言模型的测试时缩放，明确三种推理机制，提出系统评估原则与可复现性要求，应用于多类基准并发布超20亿条推理轨迹。

链接：https://arxiv.org/abs/2608.04001

作者：Mohsen Hariri, Weicong Chen, Nahal Shahini, Vikash Singh, Kai Ye, Amirhossein Samandar, Debargha Ganguly, Sreehari Sankar, Yanyan Zhang, Shouren Wang, Jerry Peng, Biyao Zhang, Michael Hinczewski, Vipin Chaudhary

英文摘要：Large language models can solve substantially harder reasoning problems with more inference-time compute. The term "test-time scaling," however, now covers diverse inference algorithms that extend deliberation along a single trajectory, sample completed candidates and aggregate them through voting or verification, or search over unfinished partial states. These algorithms differ in their statistical structure, compute accounting, and failure modes. Treating these procedures as interchangeable under a single scalar "budget," or reporting accuracy without the inference protocol that produced it, makes results difficult to compare across studies. We develop a systematic account of test-time scaling along three axes. First, we formalize test-time scaling as budgeted inference over the implicit prefix tree of an autoregressive model and distinguish three structural regimes: single-trajectory sequential scaling, leaf-level scaling with terminal reduction, and prefix-level scaling. Second, we treat the evaluated object as the entire inference system and develop evaluation principles that separate end-to-end system performance from candidate-bank diagnostics. We introduce an evaluation profile whose coordinates and simple functionals recover or bound common repeated-sampling metrics, and prescribe protocol-matched reporting of compute and uncertainty. Third, we specify reproducibility requirements for inference protocols, distinguishing exact replay from distributional reproducibility and identifying the artifacts needed to support each. We also organize the open-weight reasoning ecosystem by model-side and interface mechanisms, apply these principles to broad-knowledge, symbolic-reasoning, and competition-mathematics benchmarks, and assemble over 2 billion full reasoning traces for release with progressively richer verifier and token-level signals.