机器学习学术速递[2.3]

点击阅读原文访问arxivdaily.com，涵盖CS|物理|数学|经济|统计|金融|生物|电气领域，更有搜索、收藏等功能！

cs.LG 方向，今日共计648篇

大模型相关(110篇)

【1】Embedding Perturbation may Better Reflect the Uncertainty in LLM Reasoning
标题：嵌入微扰可能更好地反映LLM推理中的不确定性
链接：https://arxiv.org/abs/2602.02427

作者：Qihao Wen,Jiahao Wang,Yang Nan,Pengfei He,Ravi Tandon,Han Xu

【2】Repurposing Protein Language Models for Latent Flow-Based Fitness Optimization
标题：基于潜在流的适应度优化的蛋白质语言模型重构
链接：https://arxiv.org/abs/2602.02425

作者：Amaru Caceres Arroyo,Lea Bogensperger,Ahmed Allam,Michael Krauthammer,Konrad Schindler,Dominik Narnhofer

【3】An Empirical Study on Noisy Data and LLM Pretraining Loss Divergence
标题：噪音数据与LLM训练前损失分歧的实证研究
链接：https://arxiv.org/abs/2602.02400

作者：Qizhen Zhang,Ankush Garg,Jakob Foerster,Niladri Chatterji,Kshitiz Malik,Mike Lewis

【4】Trust by Design: Skill Profiles for Transparent, Cost-Aware LLM Routing
标题：设计的信任：透明的技能配置文件，成本意识LLM路由
链接：https://arxiv.org/abs/2602.02386

作者：Mika Okamoto,Ansel Kaplan Erol,Glenn Matlin
备注：Appeared at MLSys YPS 2025

【5】ReasonCACHE: Teaching LLMs To Reason Without Weight Updates
标题：ReasonCAGER：教法学硕士在没有体重的情况下推理更新
链接：https://arxiv.org/abs/2602.02366

作者：Sharut Gupta,Phillip Isola,Stefanie Jegelka,David Lopez-Paz,Kartik Ahuja,Mark Ibrahim,Mohammad Pezeshki
备注：26 pages, 17 Figures

【6】Why Steering Works: Toward a Unified View of Language Model Parameter Dynamics
标题：为什么转向有效：走向语言模型参数动态的统一视图
链接：https://arxiv.org/abs/2602.02343

作者：Ziwen Xu,Chenyan Wu,Hengyu Sun,Haiwen Hong,Mengru Wang,Yunzhi Yao,Longtao Huang,Hui Xue,Shumin Deng,Zhixuan Chu,Huajun Chen,Ningyu Zhang
备注：Work in progress

【7】Interpreting and Controlling LLM Reasoning through Integrated Policy Gradient
标题：通过综合政策梯度解释和控制LLM推理
链接：https://arxiv.org/abs/2602.02313

作者：Changming Li,Kaixing Zhang,Haoyun Xu,Yingdong Shi,Zheng Zhang,Kaitao Song,Kan Ren

【8】Position: Explaining Behavioral Shifts in Large Language Models Requires a Comparative Approach
标题：立场：解释大型语言模型中的行为转变需要比较方法
链接：https://arxiv.org/abs/2602.02304

作者：Martino Ciaperoni,Marzio Di Vece,Luca Pappalardo,Fosca Giannotti,Francesco Giannini

【9】EvalQReason: A Framework for Step-Level Reasoning Evaluation in Large Language Models
标题：EvalQReason：大型语言模型中分步推理评估框架
链接：https://arxiv.org/abs/2602.02295

作者：Shaima Ahmad Freja,Ferhat Ozgur Catak,Betul Yurdem,Chunming Rong
备注：15 pages (including appendix), 11 figures

【10】RACA: Representation-Aware Coverage Criteria for LLM Safety Testing
标题：RACA：LLM安全测试的代表感知覆盖标准
链接：https://arxiv.org/abs/2602.02280

作者：Zeming Wei,Zhixin Zhang,Chengcan Wu,Yihao Zhang,Xiaokun Luan,Meng Sun

【11】Hierarchical Adaptive Eviction for KV Cache Management in Multimodal Language Models
标题：多模式语言模型中KV缓存管理的分层自适应驱逐
链接：https://arxiv.org/abs/2602.02197

作者：Xindian Ma,Yidi Lu,Peng Zhang,Jing Zhang
备注：10 oages, 3 figures

【12】State Rank Dynamics in Linear Attention LLMs
标题：线性注意力LL中的状态等级动态
链接：https://arxiv.org/abs/2602.02195

作者：Ao Sun,Hongtao Zhang,Heng Zhou,Yixuan Ma,Yiran Qin,Tongrui Su,Yan Liu,Zhanyu Ma,Jun Xu,Jiuchong Gao,Jinghua Hao,Renqing He

【13】Vision-DeepResearch Benchmark: Rethinking Visual and Textual Search for Multimodal Large Language Models
标题：Vision-DeepResearch基准：重新思考多模式大型语言模型的视觉和文本搜索
链接：https://arxiv.org/abs/2602.02185

作者：Yu Zeng,Wenxuan Huang,Zhen Fang,Shuang Chen,Yufan Shen,Yishuo Cai,Xiaoman Wang,Zhenfei Yin,Lin Chen,Zehui Chen,Shiting Huang,Yiming Zhao,Yao Hu,Philip Torr,Wanli Ouyang,Shaosheng Cao

【14】STILL: Selecting Tokens for Intra-Layer Hybrid Attention to Linearize LLMs
标题：仍然：选择用于层内混合注意力的令牌以线性化LLM
链接：https://arxiv.org/abs/2602.02180

作者：Weikang Meng,Liangyu Huo,Yadan Luo,Jiawen Guan,Jingyi Zhang,Yingjian Li,Zheng Zhang

【15】Co-RedTeam: Orchestrated Security Discovery and Exploitation with LLM Agents
标题：联合红团队：与LLM代理一起进行安全发现和利用
链接：https://arxiv.org/abs/2602.02164

作者：Pengfei He,Ash Fox,Lesly Miculicich,Stefan Friedli,Daniel Fabian,Burak Gokturk,Jiliang Tang,Chen-Yu Lee,Tomas Pfister,Long T. Le

【16】Revisiting Adaptive Rounding with Vectorized Reparameterization for LLM Quantization
标题：重新审视LLM量化的自适应四舍五入
链接：https://arxiv.org/abs/2602.02151

作者：Yuli Zhou,Qingxuan Chen,Luca Benini,Guolei Sun,Yawei Li
备注：17 pages, 6 figures, 14 tables

【17】Two-Stage Grid Optimization for Group-wise Quantization of LLMs
标题：LLM分组量化的两阶段网格优化
链接：https://arxiv.org/abs/2602.02126

作者：Junhan Kim,Gukryeol Lee,Seungwoo Son,Jeewook Kim,Yongkweon Jeon
备注：ICASSP 2026

【18】No Global Plan in Chain-of-Thought: Uncover the Latent Planning Horizon of LLMs
标题：思想链中没有全球计划：揭开法学硕士的潜在规划视野
链接：https://arxiv.org/abs/2602.02103

作者：Liyan Xu,Mo Yu,Fandong Meng,Jie Zhou

【19】Learning to Route and Schedule LLMs from User Retrials via Contextual Queueing Bandits
标题：学习通过上下文排队盗贼从用户重审中路由和安排LLM
链接：https://arxiv.org/abs/2602.02061

作者：Seoungbin Bae,Junyoung Son,Dabeen Lee

【20】Dissecting Outlier Dynamics in LLM NVFP4 Pretraining
标题：剖析LLM NVFP 4预训练中的离群动态
链接：https://arxiv.org/abs/2602.02047

作者：Peijie Dong,Ruibo Fan,Yuechen Tao,Di Mou,Wenhu Hu,Zhenheng Tang,Yinghao Yu,Jiamang Wang,Wenbo Su,Guodong Yang,Liping Zhang,Xiaowen Chu,Baochun Li,Bo Li
备注：39 pages, 32 figures

【21】Hunt Instead of Wait: Evaluating Deep Data Research on Large Language Models
标题：寻找而不是等待：评估大型语言模型的深度数据研究
链接：https://arxiv.org/abs/2602.02039

作者：Wei Liu,Peijie Yu,Michele Orini,Yali Du,Yulan He
备注：14 pages, 7 tables, 8 figures

【22】Light Alignment Improves LLM Safety via Model Self-Reflection with a Single Neuron
标题：光线对齐通过单神经元的模型自反射提高LLM安全性
链接：https://arxiv.org/abs/2602.02027

作者：Sicheng Shen,Mingyang Lv,Han Shen,Jialin Wu,Binghao Wang,Zhou Yang,Guobin Shen,Dongcheng Zhao,Feifei Zhao,Yi Zeng
备注：21 pages, 3 figures

【23】Preserve-Then-Quantize: Balancing Rank Budgets for Quantization Error Reconstruction in LLMs
标题：保留然后量化：平衡LLM中量化误差重建的排序预算
链接：https://arxiv.org/abs/2602.02001

作者：Yoonjun Cho,Dongjae Jeon,Soeun Kim,Moongyu Jeon,Albert No

【24】On the Limits of Layer Pruning for Generative Reasoning in LLMs
标题：LLM中生成推理的层修剪限制
链接：https://arxiv.org/abs/2602.01997

作者：Safal Shrestha,Anubhav Shrestha,Aadim Nepal,Minwu Kim,Keith Ross

【25】IntraSlice: Towards High-Performance Structural Pruning with Block-Intra PCA for LLMs
标题：IntraSlice：利用LLM的块内PCA实现高性能结构修剪
链接：https://arxiv.org/abs/2602.01975

作者：Meng Li,Peisong Wang,Yuantian Shao,Qinghao Hu,Hongjian Fang,Yifan Zhang,Zhihui Wei,Jian Cheng

【26】Efficient Epistemic Uncertainty Estimation for Large Language Models via Knowledge Distillation
标题：通过知识提炼对大型语言模型进行有效的认识不确定性估计
链接：https://arxiv.org/abs/2602.01956

作者：Seonghyeon Park,Jewon Yeom,Jaewon Sok,Jeongjae Park,Heejun Kim,Taesup Kim

【27】T-LLM: Teaching Large Language Models to Forecast Time Series via Temporal Distillation
标题：T-LLM：通过时间蒸馏教授大型语言模型预测时间序列
链接：https://arxiv.org/abs/2602.01937

作者：Suhan Guo,Bingxu Wang,Shaodan Zhang,Furao Shen

【28】COLT: Lightweight Multi-LLM Collaboration through Shared MCTS Reasoning for Model Compilation
标题：COLT：通过共享MCTS推理进行模型编译的轻量级多LLM协作
链接：https://arxiv.org/abs/2602.01935

作者：Annabelle Sujun Tang,Christopher Priebe,Lianhui Qin,Hadi Esmaeilzadeh

【29】Towards Long-Horizon Interpretability: Efficient and Faithful Multi-Token Attribution for Reasoning LLMs
标题：迈向长期可解释性：推理LLM的高效且忠实的多令牌归因
链接：https://arxiv.org/abs/2602.01914

作者：Wenbo Pan,Zhichao Liu,Xianlong Wang,Haining Yu,Xiaohua Jia
备注：ICML 2025 submission

【30】Internal Flow Signatures for Self-Checking and Refinement in LLMs
标题：用于LLC中自我检查和细化的内部流程签名
链接：https://arxiv.org/abs/2602.01897

作者：Sungheon Jeong,Sanggeon Yun,Ryozo Masukawa,Wenjun Haung,Hanning Chen,Mohsen Imani

【31】Self-Rewarding Sequential Monte Carlo for Masked Diffusion Language Models
标题：掩蔽扩散语言模型的自我奖励顺序蒙特卡罗
链接：https://arxiv.org/abs/2602.01849

作者：Ziwei Luo,Ziqi Jin,Lei Wang,Lidong Bing,Thomas B. Schön
备注：Project page: https://algolzw.github.io/sr-smc

【32】No Generation without Representation: Efficient Causal Protein Language Models Enable Zero-Shot Fitness Estimation
标题：没有表示就没有生成：高效的因果蛋白质语言模型实现Zero-Shot适应度估计
链接：https://arxiv.org/abs/2602.01845

作者：Furkan Eris

【33】Prism: Efficient Test-Time Scaling via Hierarchical Search and Self-Verification for Discrete Diffusion Language Models
标题：Prism：通过离散扩散语言模型的分层搜索和自我验证的高效测试时间缩放
链接：https://arxiv.org/abs/2602.01842

作者：Jinbin Bai,Yixuan Li,Yuchen Zhu,Yi Xin,Qingyu Shi,Aosong Feng,Xiaohong Liu,Molei Tao,Jianru Xue,Xiangtai Li,Ming-Hsuan Yang

【34】Sentence Curve Language Models
标题：句子曲线语言模型
链接：https://arxiv.org/abs/2602.01807

作者：DongNyeong Heo,Heelyoul Choi

【35】Grad2Reward: From Sparse Judgment to Dense Rewards for Improving Open-Ended LLM Reasoning
标题：Grad2奖励：从稀疏判断到密集奖励，以改善开放式LLM推理
链接：https://arxiv.org/abs/2602.01791

作者：Zheng Zhang,Ao Lu,Yuanhao Zeng,Ziwei Shan,Jinjin Guo,Lufei Li,Yexin Li,Kan Ren

【36】MSign: An Optimizer Preventing Training Instability in Large Language Models via Stable Rank Restoration
标题：Mign：通过稳定的排名恢复防止大型语言模型中的训练不稳定性的优化器
链接：https://arxiv.org/abs/2602.01734

作者：Lianhai Ren,Yucheng Ding,Xiao Liu,Qianxiao Li,Peng Cheng,Yeyun Gong

【37】Cross-Domain Fake News Detection on Unseen Domains via LLM-Based Domain-Aware User Modeling
标题：通过基于LLM的领域感知用户建模在不可见的领域上进行跨领域假新闻检测
链接：https://arxiv.org/abs/2602.01726

作者：Xuankai Yang,Yan Wang,Jiajie Zhu,Pengfei Ding,Hongyang Liu,Xiuzhen Zhang,Huan Liu
备注：This paper has been accepted by The 2026 ACM Web Conference (WWW 2026)

【38】Optimizing Prompts for Large Language Models: A Causal Approach
标题：优化大型语言模型的预算：因果方法
链接：https://arxiv.org/abs/2602.01711

作者：Wei Chen,Yanbin Fang,Shuran Fu,Fasheng Xu,Xuan Wei

【39】$\textbf{AGT$^{AO}$}$: Robust and Stabilized LLM Unlearning via Adversarial Gating Training with Adaptive Orthogonality
标题：$ extBF{AGT$^{AO}$}：通过具有自适应性的对抗门控训练实现稳健且稳定的LLM去学习
链接：https://arxiv.org/abs/2602.01703

作者：Pengyu Li,Lingling Zhang,Zhitao Gao,Yanrui Wu,Yuxuan Dong,Huan Liu,Bifan Wei,Jun Liu

【40】What LLMs Think When You Don't Tell Them What to Think About?
标题：当你不告诉法学硕士该想什么时，他们会怎么想？
链接：https://arxiv.org/abs/2602.01689

作者：Yongchan Kwon,James Zou
备注：NA

【41】Semantic-aware Wasserstein Policy Regularization for Large Language Model Alignment
标题：用于大型语言模型对齐的语义感知Wasserstein政策正规化
链接：https://arxiv.org/abs/2602.01685

作者：Byeonghu Na,Hyungho Na,Yeongmin Kim,Suhyeon Jo,HeeSun Bae,Mina Kang,Il-Chul Moon
备注：Accepted at ICLR 2026

【42】Chance-Constrained Inference for Hallucination Risk Control in Large Language Models
标题：大型语言模型中幻觉风险控制的机会约束推理
链接：https://arxiv.org/abs/2602.01637

作者：Sreenivasan Mohandas

【43】A Practical Tensor-Network Compression Pipeline for Production-Scale Large Language Models
标题：用于生产规模大型语言模型的实用张量网络压缩管道
链接：https://arxiv.org/abs/2602.01613

作者：Sergii Kozyrev,Davyd Maiboroda
备注：13 pages, 5 figures

【44】Expected Harm: Rethinking Safety Evaluation of (Mis)Aligned LLMs
标题：预期危害：重新思考（不）对齐的LLM的安全性评价
链接：https://arxiv.org/abs/2602.01600

作者：Yen-Shan Chen,Zhi Rui Tam,Cheng-Kuang Wu,Yun-Nung Chen

【45】Nearly Optimal Active Preference Learning and Its Application to LLM Alignment
标题：近最优主动偏好学习及其在LLM对齐中的应用
链接：https://arxiv.org/abs/2602.01581

作者：Yao Zhao,Kwang-Sung Jun

【46】How Implicit Bias Accumulates and Propagates in LLM Long-term Memory
标题：内隐偏差在LLM长时记忆中的积累和扩展
链接：https://arxiv.org/abs/2602.01558

作者：Yiming Ma,Lixu Wang,Lionel Z. Wang,Hongkun Yang,Haoming Sun,Xin Xu,Jiaqi Wu,Bin Chen,Wei Dong
备注：Under review, and the first two authors contribute equally

【47】MAGIC: A Co-Evolving Attacker-Defender Adversarial Game for Robust LLM Safety
标题：MAGIC：一款协同进化的攻击者-防御者对抗游戏，以实现稳健的LLM安全性
链接：https://arxiv.org/abs/2602.01539

作者：Xiaoyu Wen,Zhida He,Han Qi,Ziyu Wan,Zhongtian Ma,Ying Wen,Tianhang Zheng,Xingcheng Xu,Chaochao Lu,Qiaosheng Zhang

【48】Making Bias Non-Predictive: Training Robust LLM Judges via Reinforcement Learning
标题：使偏见变得非预测性：通过强化学习训练稳健的LLM评委
链接：https://arxiv.org/abs/2602.01528

作者：Qian Wang,Xuandong Zhao,Zirui Zhang,Zhanzhi Lou,Nuo Chen,Dawn Song,Bingsheng He

【49】A Relative-Budget Theory for Reinforcement Learning with Verifiable Rewards in Large Language Model Reasoning
标题：大型语言模型推理中具有可验证奖励的强化学习相对预算理论
链接：https://arxiv.org/abs/2602.01523

作者：Akifumi Wachi,Hirota Kinoshita,Shokichi Takakura,Rei Higuchi,Taiji Suzuki
备注：28 pages

【50】Alternating Reinforcement Learning for Rubric-Based Reward Modeling in Non-Verifiable LLM Post-Training
标题：非可验证LLM后训练中基于规则的奖励模型的交替强化学习
链接：https://arxiv.org/abs/2602.01511

作者：Ran Xu,Tianci Liu,Zihan Dong,Tony You,Ilgee Hong,Carl Yang,Linjun Zhang,Tao Zhao,Haoyu Wang
备注：The first two authors contributed equally

【51】OpInf-LLM: Parametric PDE Solving with LLMs via Operator Inference
标题：OpInf-LLM：通过操作员推理使用LLM进行参数化PDL求解
链接：https://arxiv.org/abs/2602.01493

作者：Zhuoyuan Wang,Hanjiang Hu,Xiyu Deng,Saviz Mowlavi,Yorie Nakahira

【52】A Meta-Knowledge-Augmented LLM Framework for Hyperparameter Optimization in Time-Series Forecasting
标题：用于时间序列预测超参数优化的元知识增强LLM框架
链接：https://arxiv.org/abs/2602.01445

作者：Ons Saadallah,Mátyás andó,Tamás Gábor Orosz

【53】Improve the Trade-off Between Watermark Strength and Speculative Sampling Efficiency for Language Models
标题：改善语言模型水印强度和推测抽样效率之间的权衡
链接：https://arxiv.org/abs/2602.01428

作者：Weiqing He,Xiang Li,Li Shen,Weijie Su,Qi Long
备注：Accepted at ICLR 2026

【54】SNIP: An Adaptive Mixed Precision Framework for Subbyte Large Language Model Training
标题：SNIP：用于子字节大型语言模型训练的自适应混合精度框架
链接：https://arxiv.org/abs/2602.01410

作者：Yunjie Pan,Yongyi Yang,Hanmei Yang,Scott Mahlke
备注：Accepted to ASPLOS 2026

【55】Your Self-Play Algorithm is Secretly an Adversarial Imitator: Understanding LLM Self-Play through the Lens of Imitation Learning
标题：你的自我游戏算法秘密地是一个对抗模仿者：通过模仿学习的角度理解LLM自我游戏
链接：https://arxiv.org/abs/2602.01357

作者：Shangzhe Li,Xuchao Zhang,Chetan Bansal,Weitong Zhang
备注：35 pages, 5 tables, 3 figures

【56】EDIS: Diagnosing LLM Reasoning via Entropy Dynamics
标题：EDIS：通过熵动力学诊断LLM推理
链接：https://arxiv.org/abs/2602.01288

作者：Chenghua Zhu,Siyan Wu,Xiangkang Zeng,Zishan Xu,Zhaolu Kang,Yifu Guo,Yuquan Lu,Junduan Huang,Guojing Zhou
备注：Under review at ICML 2026

【57】Multi-LLM Adaptive Conformal Inference for Reliable LLM Responses
标题：可靠LLM响应的多LLM自适应共形推理
链接：https://arxiv.org/abs/2602.01285

作者：Kangjun Noh,Seongchan Lee,Ilmun Kim,Kyungwoo Song
备注：Accepted to ICLR 2026

【58】Lotus: Efficient LLM Training by Randomized Low-Rank Gradient Projection with Adaptive Subspace Switching
标题：Lotus：通过随机低等级梯度投影和自适应子空间切换进行高效LLM训练
链接：https://arxiv.org/abs/2602.01233

作者：Tianhao Miao,Zhongyuan Bao,Lejun Zhang

【59】Self-Generative Adversarial Fine-Tuning for Large Language Models
标题：大型语言模型的自生成对抗微调
链接：https://arxiv.org/abs/2602.01137

作者：Shiguang Wu,Yaqing Wang,Quanming Yao

【60】Tangent Space Fine-Tuning for Directional Preference Alignment in Large Language Models
标题：大型语言模型中方向偏好对齐的切空间微调
链接：https://arxiv.org/abs/2602.01128

作者：Mete Erdogan

【61】SPELL: Synthesis of Programmatic Edits using LLMs
标题：SpeLL：使用LLM合成程序编辑
链接：https://arxiv.org/abs/2602.01107

作者：Daniel Ramos,Catarina Gamboa,Inês Lynce,Vasco Manquinho,Ruben Martins,Claire Le Goues
备注：pre-print

【62】LRAgent: Efficient KV Cache Sharing for Multi-LoRA LLM Agents
标题：LRAgent：多LoRA LLM代理的高效KV缓存共享
链接：https://arxiv.org/abs/2602.01053

作者：Hyesung Jeon,Hyeongju Ha,Jae-Joon Kim
备注：23 pages, 9 figures, 19 tables

【63】SFMP: Fine-Grained, Hardware-Friendly and Search-Free Mixed-Precision Quantization for Large Language Models
标题：SFMP：针对大型语言模型的细粒度、硬件友好且免搜索的混合精度量化
链接：https://arxiv.org/abs/2602.01027

作者：Xin Nie,Haicheng Zhang,Liang Dong,Beining Feng,Jinhong Weng,Guiling Sun
备注：24pages,17figures

【64】Toward Universal and Transferable Jailbreak Attacks on Vision-Language Models
标题：走向对视觉语言模型的通用和可转移越狱攻击
链接：https://arxiv.org/abs/2602.01025

作者：Kaiyuan Cui,Yige Li,Yutao Wu,Xingjun Ma,Sarah Erfani,Christopher Leckie,Hanxun Huang
备注：ICLR 2026

【65】ESSAM: A Novel Competitive Evolution Strategies Approach to Reinforcement Learning for Memory Efficient LLMs Fine-Tuning
标题：ESSam：一种新型的竞争进化策略方法，用于记忆高效的LLM微调
链接：https://arxiv.org/abs/2602.01003

作者：Zhishen Sun,Sizhe Dang,Guang Dai,Haishan Ye

【66】DISPO: Enhancing Training Efficiency and Stability in Reinforcement Learning for Large Language Model Mathematical Reasoning
标题：DISPO：提高大型语言模型数学推理强化学习的训练效率和稳定性
链接：https://arxiv.org/abs/2602.00983

作者：Batuhan K. Karaman,Aditya Rawal,Suhaila Shakiah,Mohammad Ghavamzadeh,Mingyi Hong,Arijit Biswas,Ruida Zhou
备注：This work is accepted to the 29th International Conference on Artificial Intelligence and Statistics (AISTATS) 2026

【67】Trust in One Round: Confidence Estimation for Large Language Models via Structural Signals
标题：一轮信任：通过结构信号对大型语言模型进行置信度估计
链接：https://arxiv.org/abs/2602.00977

作者：Pengyue Yang,Jiawen Wen,Haolin Jin,Linghan Huang,Huaming Chen,Ling Chen
备注：Accepted at The ACM Web Conference 2026 (WWW 2026)

【68】Optimal Budgeted Adaptation of Large Language Models
标题：大型语言模型的最佳预适应
链接：https://arxiv.org/abs/2602.00952

作者：Jing Wang,Jie Shen,Dean Foster,Zohar Karnin,Jeremy C Weiss

【69】EffGen: Enabling Small Language Models as Capable Autonomous Agents
标题：EffGen：使小型语言模型成为有能力的自治代理
链接：https://arxiv.org/abs/2602.00887

作者：Gaurav Srivastava,Aafiya Hussain,Chi Wang,Yingyan Celine Lin,Xuan Wang

【70】Reliability-Aware Determinantal Point Processes for Robust Informative Data Selection in Large Language Models
标题：用于大型语言模型中稳健信息数据选择的可靠性感知决定点过程
链接：https://arxiv.org/abs/2602.00885

作者：Ahmad Sarlak,Abolfazl Razi

【71】Dynamic Expert Sharing: Decoupling Memory from Parallelism in Mixture-of-Experts Diffusion LLMs
标题：动态专家共享：在混合专家扩散LLM中将记忆与平行主义脱钩
链接：https://arxiv.org/abs/2602.00879

作者：Hao Mark Chen,Zhiwen Mo,Royson Lee,Qianzhou Wang,Da Li,Shell Xu Hu,Wayne Luk,Timothy Hospedales,Hongxiang Fan

【72】Provable Model Provenance Set for Large Language Models
标题：大型语言模型的可证明模型出处集
链接：https://arxiv.org/abs/2602.00772

作者：Xiaoqi Qiu,Hao Zeng,Zhiyu Hou,Hongxin Wei

【73】Provably Protecting Fine-Tuned LLMs from Training Data Extraction
标题：可证明保护精调LLM免受训练数据提取的影响
链接：https://arxiv.org/abs/2602.00688

作者：Tom Segal,Asaf Shabtai,Yuval Elovici
备注：20 pages, 5 figures

【74】SEISMO: Increasing Sample Efficiency in Molecular Optimization with a Trajectory-Aware LLM Agent
标题：SEISMO：使用轨迹感知LLM代理提高分子优化中的样本效率
链接：https://arxiv.org/abs/2602.00663

作者：Fabian P. Krüger,Andrea Hunklinger,Adrian Wolny,Tim J. Adler,Igor Tetko,Santiago David Villalba
备注：Fabian P. Krüger and Andrea Hunklinger contributed equally to this work

【75】From Associations to Activations: Comparing Behavioral and Hidden-State Semantic Geometry in LLMs
标题：从关联到激活：LLM中的行为和隐藏状态语义几何的比较
链接：https://arxiv.org/abs/2602.00628

作者：Louis Schiekiera,Max Zimmer,Christophe Roux,Sebastian Pokutta,Fritz Günther
备注：25 pages including references, 15 figures, 6 tables

【76】Jailbreaking LLMs via Calibration
标题：通过校准越狱LLM
链接：https://arxiv.org/abs/2602.00619

作者：Yuxuan Lu,Yongkang Guo,Yuqing Kong

【77】Sparsity-Aware Unlearning for Large Language Models
标题：大型语言模型的空间意识取消学习
链接：https://arxiv.org/abs/2602.00577

作者：Yuze Wang,Yujia Tong,Ke Xu,Jingling Yuan,Jiawei Jiang,Chuang Hu

【78】Data Distribution as a Lever for Guiding Optimizers Toward Superior Generalization in LLMs
标题：数据分布作为引导优化者在LLM中实现卓越概括的杠杆
链接：https://arxiv.org/abs/2602.00576

作者：Tushaar Gangavarapu,Jiping Li,Christopher Vattheuer,Zhangyang Wang,Baharan Mirzasoleiman

【79】Minerva: Reinforcement Learning with Verifiable Rewards for Cyber Threat Intelligence LLMs
标题：Minerva：网络威胁情报LLM的强化学习和可验证奖励
链接：https://arxiv.org/abs/2602.00513

作者：Md Tanvirul Alam,Aritran Piplai,Ionut Cardei,Nidhi Rastogi,Peter J Worth

【80】PCBSchemaGen: Constraint-Guided Schematic Design via LLM for Printed Circuit Boards (PCB)
标题：PCBSchemaGen：通过LLM进行印刷电路板（PCB）的约束引导原理图设计
链接：https://arxiv.org/abs/2602.00510

作者：Huanghaohe Zou,Peng Han,Emad Nazerian,Alex Q. Huang

【81】AREAL-DTA: Dynamic Tree Attention for Efficient Reinforcement Learning of Large Language Models
标题：AREAL-DART：动态树注意力，用于大型语言模型的高效强化学习
链接：https://arxiv.org/abs/2602.00482

作者：Jiarui Zhang,Yuchen Yang,Ran Yan,Zhiyu Mei,Liyuan Zhang,Daifeng Li,Wei Fu,Jiaxuan Gao,Shusheng Xu,Yi Wu,Binhang Yuan

【82】FedMOA: Federated GRPO for Personalized Reasoning LLMs under Heterogeneous Rewards
标题：FedMOA：用于异类奖励下的个性化推理LLM的联合GRPO
链接：https://arxiv.org/abs/2602.00453

作者：Ziyao Wang,Daeun Jung,Yexiao He,Guoheng Sun,Zheyu Shen,Myungjin Lee,Ang Li

【83】LLMs as High-Dimensional Nonlinear Autoregressive Models with Attention: Training, Alignment and Inference
标题：LLM作为具有注意力的多维非线性自回归模型：训练、对齐和推理
链接：https://arxiv.org/abs/2602.00426

作者：Vikram Krishnamurthy
备注：27 pages, 12 figures. Mathematical survey framing LLMs as high-dimensional nonlinear autoregressive models with attention, covering training, alignment, and inference, with nanoGPT/nanochat-style code examples. Feedback welcome

【84】Toward Autonomous Laboratory Safety Monitoring with Vision Language Models: Learning to See Hazards Through Scene Structure
标题：使用视觉语言模型实现自主实验室安全监控：学会通过场景结构看到危险
链接：https://arxiv.org/abs/2602.00414

作者：Trishna Chakraborty,Udita Ghosh,Aldair Ernesto Gongora,Ruben Glatt,Yue Dong,Jiachen Li,Amit K. Roy-Chowdhury,Chengyu Song

【85】Fast Forward: Accelerating LLM Prefill with Predictive FFN Sparsity
标题：快进：使用预测性FFN稀疏性加速LLM预填充
链接：https://arxiv.org/abs/2602.00397

作者：Aayush Gautam,Mukul Gagrani,Junyoung Park,Mingu Lee,Chiris Lott,Narasimha Reddy
备注：10 pages, 7 figures

【86】A Fragile Guardrail: Diffusion LLM's Safety Blessing and Its Failure Mode
标题：脆弱的守护者：扩散LLM的安全祝福及其失败模式
链接：https://arxiv.org/abs/2602.00388

作者：Zeyuan He,Yupeng Chen,Lang Lin,Yihan Wang,Shenxu Chang,Eric Sommerlade,Philip Torr,Junchi Yu,Adel Bibi,Jialin Yu

【87】Planning with Language and Generative Models: Toward General Reward-Guided Wireless Network Design
标题：使用语言和生成模型进行规划：走向一般奖励引导的无线网络设计
链接：https://arxiv.org/abs/2602.00357

作者：Chenyang Yuan,Xiaoyuan Cheng

【88】Efficient and accurate steering of Large Language Models through attention-guided feature learning
标题：通过注意力引导特征学习高效准确地引导大型语言模型
链接：https://arxiv.org/abs/2602.00333

作者：Parmida Davarmanesh,Ashia Wilson,Adityanarayanan Radhakrishnan

【89】Harvest: Opportunistic Peer-to-Peer GPU Caching for LLM Inference
标题：Harvest：用于LLM推理的启发式对等图形处理器
链接：https://arxiv.org/abs/2602.00328

作者：Nikhil Gopal,Kostis Kaffes

【90】Semantics-Preserving Evasion of LLM Vulnerability Detectors
标题：LLM漏洞检测器的保留语义规避
链接：https://arxiv.org/abs/2602.00305

作者：Luze Sun,Alina Oprea,Eric Wong

【91】Benchmarking Uncertainty Calibration in Large Language Model Long-Form Question Answering
标题：大型语言模型长式问题回答中的不确定性校准基准
链接：https://arxiv.org/abs/2602.00279

作者：Philip Müller,Nicholas Popovič,Michael Färber,Peter Steinbach
备注：Under Review

【92】VoxServe: Streaming-Centric Serving System for Speech Language Models
标题：VoxServe：以流媒体为中心的语音语言模型服务系统
链接：https://arxiv.org/abs/2602.00269

作者：Keisuke Kamahori,Wei-Tzu Lee,Atindra Jha,Rohan Kadekodi,Stephanie Wang,Arvind Krishnamurthy,Baris Kasikci
备注：The code is available at https://github.com/vox-serve/vox-serve

【93】CAPA: Contribution-Aware Pruning and FFN Approximation for Efficient Large Vision-Language Models
标题：APA：高效大型视觉语言模型的贡献感知修剪和FFN逼近
链接：https://arxiv.org/abs/2602.00247

作者：Samyak Jha,Junho Kim

【94】Dispersion Loss Counteracts Embedding Condensation and Improves Generalization in Small Language Models
标题：分散损失抵消了嵌入式浓缩并改进了小型语言模型中的概括
链接：https://arxiv.org/abs/2602.00217

作者：Chen Liu,Xingzhi Sun,Xi Xiao,Alexandre Van Tassel,Ke Xu,Kristof Reimann,Danqi Liao,Mark Gerstein,Tianyang Wang,Xiao Wang,Smita Krishnaswamy

【95】From Gameplay Traces to Game Mechanics: Causal Induction with Large Language Models
标题：从游戏轨迹到游戏机制：大型语言模型的因果推理
链接：https://arxiv.org/abs/2602.00190

作者：Mohit Jiwatode,Alexander Dockhorn,Bodo Rosenhahn
备注：Submitted to ICPR 2026

【96】The Blessing of Dimensionality in LLM Fine-tuning: A Variance-Curvature Perspective
标题：LLM微调中专业性的祝福：方差-曲线的视角
链接：https://arxiv.org/abs/2602.00170

作者：Qiyao Liang,Jinyeop Song,Yizhou Liu,Jeff Gore,Ila Fiete,Risto Miikkulainen,Xin Qiu
备注：8 pages, 6 figures, plus appendices

【97】Joint Continual Learning of Local Language Models and Cloud Offloading Decisions with Budget Constraints
标题：本地语言模型的联合持续学习和具有预算限制的云卸载决策
链接：https://arxiv.org/abs/2602.00166

作者：Evan Chen,Wenzhi Fang,Shiqiang Wang,Christopher Brinton

【98】Benford's Law as a Distributional Prior for Post-Training Quantization of Large Language Models
标题：本福德定律作为大型语言模型训练后量化的分布先验
链接：https://arxiv.org/abs/2602.00165

作者：Arthur Negrão,Pedro Silva,Vander L. S. Freitas,Gladston Moreira,Eduardo Luz

【99】Block removal for large language models through constrained binary optimization
标题：通过约束二进制优化对大型语言模型进行块删除
链接：https://arxiv.org/abs/2602.00161

作者：David Jansen,Roman Rausch,David Montero,Roman Orus
备注：7 pages, 5 figures

【100】Monte Carlo Tree Search for Execution-Guided Program Repair with Large Language Models
标题：使用大型语言模型进行执行引导程序修复的蒙特卡洛树搜索
链接：https://arxiv.org/abs/2602.00129

作者：Yixuan Liang
备注：10 pages, 5 figures. Submitted to a conference workshop

【101】ALIGN: Aligned Delegation with Performance Guarantees for Multi-Agent LLM Reasoning
标题：ALIGN：多代理LLM推理的一致委托与性能保证
链接：https://arxiv.org/abs/2602.00127

作者：Tong Zhu,Baiting Chen,Jin Zhou,Hua Zhou,Sriram Sankararaman,Xiaowu Dai

【102】CARE-RFT: Confidence-Anchored Reinforcement Finetuning for Reliable Reasoning in Large Language Models
标题：CARE-RFT：在大型语言模型中进行可靠推理的信任锚定强化微调
链接：https://arxiv.org/abs/2602.00085

作者：Shuozhe Li,Jincheng Cao,Bodun Hu,Aryan Mokhtari,Leqi Liu,Amy Zhang

【103】FoundationalASSIST: An Educational Dataset for Foundational Knowledge Tracing and Pedagogical Grounding of LLMs
标题：FoundationalASSIST：用于LLM基础知识追踪和教学基础的教育数据集
链接：https://arxiv.org/abs/2602.00070

作者：Eamon Worden,Cristina Heffernan,Neil Heffernan,Shashank Sonkar

【104】Enhancing few-shot time series forecasting with LLM-guided diffusion
标题：利用LLM引导的扩散增强Few-Shot时间序列预测
链接：https://arxiv.org/abs/2602.00040

作者：Haonan Shi,Dehua Shuai,Liming Wang,Xiyang Liu,Long Tian

【105】ELLMPEG: An Edge-based Agentic LLM Video Processing Tool
标题：ELLJPEG：一个基于边缘的统计LLM视频处理工具
链接：https://arxiv.org/abs/2602.00028

作者：Zoha Azimi,Reza Farahani,Radu Prodan,Christian Timmerer
备注：12 pages, 5 tables, 8 Figures, accepted for the MMSys 2026 conference

【106】OGD4All: A Framework for Accessible Interaction with Geospatial Open Government Data Based on Large Language Models
标题：OGD 4All：基于大型语言模型的地理空间开放政府数据可访问交互的框架
链接：https://arxiv.org/abs/2602.00012

作者：Michael Siebenmann,Javier Argota Sánchez-Vaquerizo,Stefan Arisona,Krystian Samp,Luis Gisler,Dirk Helbing
备注：This work has been submitted to the IEEE for possible publication. 7 pages, 6 figures

【107】C$^2$-Cite: Contextual-Aware Citation Generation for Attributed Large Language Models
标题：C$#2 $-Cite：归因大型语言模型的上下文感知引文生成
链接：https://arxiv.org/abs/2602.00004

作者：Yue Yu,Ting Bai,HengZhi Lan,Li Qian,Li Peng,Jie Wu,Wei Liu,Jian Luan,Chuan Shi
备注：WSDM26

【108】Efficient Multilingual Search Relevance Modeling in E-Commerce via LLM Mixture-of-Experts
标题：通过LLM专家混合进行电子商务中的高效多语言搜索相关性建模
链接：https://arxiv.org/abs/2602.00003

作者：Ye Liu,Xu Chen,Wuji Chen,Mang Li
备注：4 pages, 2 figures

【109】Inference-Aware Meta-Alignment of LLMs via Non-Linear GRPO
标题：通过非线性GRPO进行LLM的推理感知元对齐
链接：https://arxiv.org/abs/2602.01603

作者：Shokichi Takakura,Akifumi Wachi,Rei Higuchi,Kohei Miyaguchi,Taiji Suzuki

【110】Alignment of Diffusion Model and Flow Matching for Text-to-Image Generation
标题：文本到图像生成的扩散模型和流匹配的匹配
链接：https://arxiv.org/abs/2602.00413

作者：Yidong Ouyang,Liyan Xie,Hongyuan Zha,Guang Cheng

Graph相关(图学习|图神经网络|图优化等)(28篇)

【1】HopFormer: Sparse Graph Transformers with Explicit Receptive Field Control
标题：HopFormer：具有显式感受场控制的稀疏图变形机
链接：https://arxiv.org/abs/2602.02268

作者：Sanggeon Yun,Raheeb Hassan,Ryozo Masukawa,Sungheon Jeong,Mohsen Imani

【2】Cardinality-Preserving Structured Sparse Graph Transformers for Molecular Property Prediction
标题：用于分子性质预测的保基数结构稀疏图变换器
链接：https://arxiv.org/abs/2602.02201

作者：Abhijit Gupta

【3】Generating Causal Temporal Interaction Graphs for Counterfactual Validation of Temporal Link Prediction
标题：生成因果时间相互作用图用于时间链接预测的反事实验证
链接：https://arxiv.org/abs/2602.02161

作者：Aniq Ur Rahman,Justin P. Coon

【4】Twinning Complex Networked Systems: Data-Driven Calibration of the mABCD Synthetic Graph Generator
标题：配对复杂网络系统：mABCD合成图生成器的数据驱动校准
链接：https://arxiv.org/abs/2602.02044

作者：Piotr Bródka,Michał Czuba,Bogumił Kamiński,Łukasz Kraiński,Katarzyna Musial,Paweł Prałat,Mateusz Stolarski

【5】PIMPC-GNN: Physics-Informed Multi-Phase Consensus Learning for Enhancing Imbalanced Node Classification in Graph Neural Networks
标题：PIMPC-GNN：物理信息多阶段共识学习，用于增强图神经网络中的不平衡节点分类
链接：https://arxiv.org/abs/2602.01920

作者：Abdul Joseph Fofanah,Lian Wen,David Chen

【6】Grappa: Gradient-Only Communication for Scalable Graph Neural Network Training
标题：Grappa：可扩展图神经网络训练的仅限参与者通信
链接：https://arxiv.org/abs/2602.01872

作者：Chongyang Xu,Christoph Siebenbrunner,Laurent Bindschaedler

【7】Hyperbolic Graph Neural Networks Under the Microscope: The Role of Geometry-Task Alignment
标题：微观下的双曲图神经网络：几何与任务对齐的作用
链接：https://arxiv.org/abs/2602.01828

作者：Dionisia Naddeo,Jonas Linkerhägner,Nicola Toschi,Geri Skenderi,Veronica Lachi

【8】MGKAN: Predicting Asymmetric Drug-Drug Interactions via a Multimodal Graph Kolmogorov-Arnold Network
标题：MGKAN：通过多峰图Kolmogorov-Arnold网络预测不对称药物相互作用
链接：https://arxiv.org/abs/2602.01751

作者：Kunyi Fan,Mengjie Chen,Longlong Li,Cunquan Qu
备注：Submitted to ICASSP 2026

【9】Modeling Topological Impact on Node Attribute Distributions in Attributed Graphs
标题：属性图中对节点属性分布的布局影响建模
链接：https://arxiv.org/abs/2602.01454

作者：Amirreza Shiralinasab Langari,Leila Yeganeh,Kim Khoa Nguyen

【10】Key Principles of Graph Machine Learning: Representation, Robustness, and Generalization
标题：图形机器学习的关键原则：表示、鲁棒性和概括
链接：https://arxiv.org/abs/2602.01139

作者：Yassine Abbahaddou
备注：PhD Thesis

【11】ChronoSpike: An Adaptive Spiking Graph Neural Network for Dynamic Graphs
标题：ChronoSpike：用于动态图形的自适应尖峰图神经网络
链接：https://arxiv.org/abs/2602.01124

作者：Md Abrar Jahin,Taufikur Rahman Fuad,Jay Pujara,Craig Knoblock

【12】Single-Edge Node Injection Threats to GNN-Based Security Monitoring in Industrial Graph Systems
标题：单边节点注入对工业图系统中基于GNN的安全监控的威胁
链接：https://arxiv.org/abs/2602.01113

作者：Wenjie Liang,Ranhui Yan,Jia Cai,You-Gan Wang

【13】GAPNet: Plug-in Jointly Learning Task-Specific Graph for Dynamic Stock Relation
标题：GAPNet：插件联合学习动态股票关系的特定任务图
链接：https://arxiv.org/abs/2602.00888

作者：Yingjie Niu,Lanxin Lu,Changhong Jin,Ruihai Dong

【14】Towards Multiscale Graph-based Protein Learning with Geometric Secondary Structural Motifs
标题：实现具有几何二级结构基元的多尺度基于图的蛋白质学习
链接：https://arxiv.org/abs/2602.00862

作者：Shih-Hsin Wang,Yuhao Huang,Taos Transue,Justin Baker,Jonathan Forstater,Thomas Strohmer,Bao Wang
备注：Published in NeurIPS 2025

【15】Sporadic Gradient Tracking over Directed Graphs: A Theoretical Perspective on Decentralized Federated Learning
标题：有向图上的零星梯度跟踪：去中心化联邦学习的理论视角
链接：https://arxiv.org/abs/2602.00791

作者：Shahryar Zehtabi,Dong-Jun Han,Seyyedali Hosseinalipour,Christopher Brinton
备注：32 pages, 5 figures

【16】GraphNNK -- Graph Classification and Interpretability
标题：GraphNNK --图分类和解释性
链接：https://arxiv.org/abs/2602.00753

作者：Zeljko Bolevic,Milos Brajovic,Isidora Stankovic,Ljubisa Stankovic
备注：4 pages, 3 figures, IEEE conference paper

【17】Non-Clashing Teaching in Graphs: Algorithms, Complexity, and Bounds
标题：图形的非冲突教学：算法、复杂性和界限
链接：https://arxiv.org/abs/2602.00657

作者：Sujoy Bhore,Liana Khazaliya,Fionn Mc Inerney
备注：An extended abstract of this paper will appear in the proceedings of ICLR 2026

【18】Riemannian Flow Matching for Disentangled Graph Domain Adaptation
标题：用于解纠缠图域自适应的Riemann流匹配
链接：https://arxiv.org/abs/2602.00656

作者：Yingxu Wang,Xinwang Liu,Mengzhu Wang,Siyang Gao,Nan Yin

【19】Kernelized Edge Attention: Addressing Semantic Attention Blurring in Temporal Graph Neural Networks
标题：核心化边缘注意力：解决时间图神经网络中的语义注意力模糊问题
链接：https://arxiv.org/abs/2602.00596

作者：Govind Waghmare,Srini Rohan Gujulla Leel,Nikhil Tumbde,Sumedh B G,Sonia Gupta,Srikanta Bedathur
备注：Accepted at AAAI 2026

【20】PolarMem: A Training-Free Polarized Latent Graph Memory for Verifiable Multimodal Agents
标题：PolarMem：可验证多模式代理的免训练极化潜图记忆
链接：https://arxiv.org/abs/2602.00415

作者：Zhisheng Chen,Tingyu Wu,Zijie Zhou,Zhengwei Xie,Ziyan Weng,Yingwei Zhang

【21】Fed-Listing: Federated Label Distribution Inference in Graph Neural Networks
标题：Fed-Listing：图神经网络中的联邦标签分布推断
链接：https://arxiv.org/abs/2602.00407

作者：Suprim Nakarmi,Junggab Son,Yue Zhao,Zuobin Xiong
备注：13 pages, 4 figures, and 5 tables

【22】DROGO: Default Representation Objective via Graph Optimization in Reinforcement Learning
标题：DROGO：强化学习中通过图优化的默认表示目标
链接：https://arxiv.org/abs/2602.00403

作者：Hon Tik Tse,Marlos C. Machado

【23】Optimal Transport-Guided Adversarial Attacks on Graph Neural Network-Based Bot Detection
标题：基于图神经网络的Bot检测的最佳传输引导对抗攻击
链接：https://arxiv.org/abs/2602.00318

作者：Kunal Mukherjee,Zulfikar Alom,Tran Gia Bao Ngo,Cuneyt Gurcan Akcora,Murat Kantarcioglu

【24】Modality as Heterogeneity: Node Splitting and Graph Rewiring for Multimodal Graph Learning
标题：作为异类的情态：多情态图学习的节点分裂和图重新布线
链接：https://arxiv.org/abs/2602.00067

作者：Yihan Zhang,Ercan E. Kuruoglu

【25】SPGCL: Effective Graph Contrastive Learning via SVD-Guided Structural Perturbation
标题：SPGCL：通过ASD引导的结构扰动进行有效的图对比学习
链接：https://arxiv.org/abs/2602.00064

作者：Hao Deng,Yingping Li,Shuiping Gou,Bo Liu

【26】A New Workflow for Materials Discovery Bridging the Gap Between Experimental Databases and Graph Neural Networks
标题：材料发现的新工作流程弥合实验数据库和图神经网络之间的差距
链接：https://arxiv.org/abs/2602.00756

作者：Brandon Schoener,Yuting Hu,Pasit Wanlapha,Akshay Rengarajan,Ian Moog,Michael Wang,Peihong Zhang,Jinjun Xiong,Hao Zeng
备注：8 pages, 3 figures, 1 table, submitted to Journal of Magnetism and Magnetic Materials

【27】RAG-GNN: Integrating Retrieved Knowledge with Graph Neural Networks for Precision Medicine
标题：RAG-GNN：将检索到的知识与图形神经网络集成用于精准医学
链接：https://arxiv.org/abs/2602.00586

作者：Hasi Hays,William J. Richardson

【28】Explore Brain-Inspired Machine Intelligence for Connecting Dots on Graphs Through Holographic Blueprint of Oscillatory Synchronization
标题：探索大脑启发的机器智能，通过振荡同步的全息蓝图连接图形上的点
链接：https://arxiv.org/abs/2602.00057

作者：Tingting Dan,Jiaqi Ding,Guorong Wu
备注：Published in Nature Communications

Transformer(18篇)

【1】Transformers learn factored representations
标题：Transformer学习分解表示
链接：https://arxiv.org/abs/2602.02385

作者：Adam Shai,Loren Amdahl-Culleton,Casper L. Christensen,Henry R. Bigelow,Fernando E. Rosas,Alexander B. Boyd,Eric A. Alt,Kyle J. Ray,Paul M. Riechers

【2】SEDformer: Event-Synchronous Spiking Transformers for Irregular Telemetry Time Series Forecasting
标题：SEDformer：用于不规则遥感时间序列预测的事件同步尖峰Transformer
链接：https://arxiv.org/abs/2602.02230

作者：Ziyu Zhou,Yuchen Fang,Weilin Ruan,Shiyu Wang,James Kwok,Yuxuan Liang
备注：Under review

【3】Time2Vec-Integrated Transformer for Robust Gesture Recognition from Low-Density sEMG
标题：Time 2Vec集成Transformer，用于从低密度sEMG中进行稳健的手势识别
链接：https://arxiv.org/abs/2602.01855

作者：Blagoj Hristov,Hristijan Gjoreski,Vesna Ojleska Latkoska,Gorjan Nadzinski

【4】Designing Time Series Experiments in A/B Testing with Transformer Reinforcement Learning
标题：利用Transformer强化学习设计A/B测试中的时间序列实验
链接：https://arxiv.org/abs/2602.01853

作者：Xiangkun Wu,Qianglin Wen,Yingying Zhang,Hongtu Zhu,Ting Li,Chengchun Shi

【5】Spatio-Temporal Transformers for Long-Term NDVI Forecasting
标题：长期诺和诺德预测的时空转换器
链接：https://arxiv.org/abs/2602.01799

作者：Ido Faran,Nathan S. Netanyahu,Maxim Shoshany

【6】CoMeT: Collaborative Memory Transformer for Efficient Long Context Modeling
标题：CoMeT：用于高效长上下文建模的协作内存Transformer
链接：https://arxiv.org/abs/2602.01766

作者：Runsong Zhao,Shilei Liu,Jiwei Tang,Langming Liu,Haibin Chen,Weidong Zhang,Yujin Yuan,Tong Xiao,Jingbo Zhu,Wenbo Su,Bo Zheng

【7】Plain Transformers are Surprisingly Powerful Link Predictors
标题：普通Transformer是令人惊讶的强大链接预测者
链接：https://arxiv.org/abs/2602.01553

作者：Quang Truong,Yu Song,Donald Loveland,Mingxuan Ju,Tong Zhao,Neil Shah,Jiliang Tang

【8】Multi-Scale Wavelet Transformers for Operator Learning of Dynamical Systems
标题：用于动态系统操作学习的多尺度子波变换器
链接：https://arxiv.org/abs/2602.01486

作者：Xuesong Wang,Michael Groom,Rafael Oliveira,He Zhao,Terence O'Kane,Edwin V. Bonilla

【9】Understanding vision transformer robustness through the lens of out-of-distribution detection
标题：通过分布外检测的视角了解视觉Transformer器的鲁棒性
链接：https://arxiv.org/abs/2602.01459

作者：Joey Kuang,Alexander Wong
备注：Accepted to JCVIS 2025

【10】TQL: Scaling Q-Functions with Transformers by Preventing Attention Collapse
标题：TQL：通过防止注意力崩溃来使用Transformer缩放Q函数
链接：https://arxiv.org/abs/2602.01439

作者：Perry Dong,Kuo-Han Hung,Alexander Swerdlow,Dorsa Sadigh,Chelsea Finn

【11】Semi-supervised CAPP Transformer Learning via Pseudo-labeling
标题：通过伪标签实现半监督的CAP Transformer学习
链接：https://arxiv.org/abs/2602.01419

作者：Dennis Gross,Helge Spieker,Arnaud Gotlieb,Emmanuel Stathatos,Panorios Benardos,George-Christopher Vosniakos

【12】SEER: Transformer-based Robust Time Series Forecasting via Automated Patch Enhancement and Replacement
标题：SEER：通过自动补丁增强和替换进行基于转换器的稳健时间序列预测
链接：https://arxiv.org/abs/2602.00589

作者：Xiangfei Qiu,Xvyuan Liu,Tianen Shen,Xingjian Wu,Hanyin Cheng,Bin Yang,Jilin Hu

【13】NEST: Nested Event Stream Transformer for Sequences of Multisets
标题：NEST：用于多集序列的嵌套事件流Transformer
链接：https://arxiv.org/abs/2602.00520

作者：Minghui Sun,Haoyu Gong,Xingyu You,Jillian Hurst,Benjamin Goldstein,Matthew Engelhard
备注：11 pages

【14】MemoryLLM: Plug-n-Play Interpretable Feed-Forward Memory for Transformers
标题：MEDoryLLM：适用于Transformer的即插即用可解释的前向存储器
链接：https://arxiv.org/abs/2602.00398

作者：Ajay Jaiswal,Lauren Hannah,Han-Byul Kim,Duc Hoang,Arnav Kundu,Mehrdad Farajtabar,Minsik Cho

【15】Brazilian Portuguese Image Captioning with Transformers: A Study on Cross-Native-Translated Dataset
标题：巴西葡萄牙语《Transformer》图像字幕：跨本地翻译数据集研究
链接：https://arxiv.org/abs/2602.00393

作者：Gabriel Bromonschenkel,Alessandro L. Koerich,Thiago M. Paixão,Hilário Tomaz Alves de Oliveira
备注：Accepted to JBCS. 18 pages, 11 figures

【16】Observing Health Outcomes Using Remote Sensing Imagery and Geo-Context Guided Visual Transformer
标题：使用遥感图像和地理背景引导视觉Transformer观察健康结果
链接：https://arxiv.org/abs/2602.00110

作者：Yu Li,Guilherme N. DeSouza,Praveen Rao,Chi-Ren Shyu
备注：Submitted to IEEE Transactions on Geoscience and Remote Sensing

【17】Transformers as Measure-Theoretic Associative Memory: A Statistical Perspective and Minimax Optimality
标题：Transformer作为测量理论联想记忆：统计视角和极小最优性
链接：https://arxiv.org/abs/2602.01863

作者：Ryotaro Kawata,Taiji Suzuki

【18】RIR-Former: Coordinate-Guided Transformer for Continuous Reconstruction of Room Impulse Responses
标题：RIR-former：用于连续重建房间冲击响应的坐标引导Transformer
链接：https://arxiv.org/abs/2602.01861

作者：Shaoheng Xu,Chunyi Sun,Jihui,Zhang,Prasanga N. Samarasinghe,Thushara D. Abhayapala
备注：Accepted to International Conference on Acoustics, Speech and Signal Processing (ICASSP) 2026. Equal contribution: Shaoheng Xu and Chunyi Sun

GAN|对抗|攻击|生成相关(26篇)

【1】Personalized Image Generation via Human-in-the-loop Bayesian Optimization
标题：通过人在环Bayesian优化实现个性化图像生成
链接：https://arxiv.org/abs/2602.02388

作者：Rajalaxmi Rajagopalan,Debottam Dutta,Yu-Lin Wei,Romit Roy Choudhury

【2】Generating Physically Sound Designs from Text and a Set of Physical Constraints
标题：从文本和一组物理约束生成物理合理的设计
链接：https://arxiv.org/abs/2602.02213

作者：Gregory Barber,Todd C. Henry,Mulugeta A. Haile
备注：NeurIPS 2025

【3】Enhancing Diffusion-Based Quantitatively Controllable Image Generation via Matrix-Form EDM and Adaptive Vicinal Training
标题：通过矩阵形式电火花加工和自适应邻位训练增强基于扩散的量化可控图像生成
链接：https://arxiv.org/abs/2602.02114

作者：Xin Ding,Yun Chen,Sen Zhang,Kao Zhang,Nenglun Chen,Peibei Cao,Yongwei Wang,Fei Wu

【4】Unifying Masked Diffusion Models with Various Generation Orders and Beyond
标题：统一具有不同世代阶数及以上的掩蔽扩散模型
链接：https://arxiv.org/abs/2602.02112

作者：Chunsan Hong,Sanghyun Lee,Jong Chul Ye
备注：Preprint

【5】Boundary-Constrained Diffusion Models for Floorplan Generation: Balancing Realism and Diversity
标题：平面设计生成的边界约束扩散模型：平衡现实主义和多样性
链接：https://arxiv.org/abs/2602.01949

作者：Leonardo Stoppani,Davide Bacciu,Shahab Mokarizadeh
备注：Accepted at ESANN 2026

【6】Zero2Text: Zero-Training Cross-Domain Inversion Attacks on Textual Embeddings
标题：Zero2text：对文本嵌入的零训练跨域翻转攻击
链接：https://arxiv.org/abs/2602.01757

作者：Doohyun Kim,Donghwa Kang,Kyungjae Lee,Hyeongboo Baek,Brent Byunghoon Kang
备注：10 pages

【7】Adversarial Reward Auditing for Active Detection and Mitigation of Reward Hacking
标题：用于主动检测和缓解奖励黑客攻击的对抗性奖励审计
链接：https://arxiv.org/abs/2602.01750

作者：Mohammad Beigi,Ming Jin,Junshan Zhang,Qifan Wang,Lifu Huang

【8】Efficient Adversarial Attacks on High-dimensional Offline Bandits
标题：对多维离线盗贼的高效对抗攻击
链接：https://arxiv.org/abs/2602.01658

作者：Seyed Mohammad Hadi Hosseini,Amir Najafi,Mahdieh Soleymani Baghshah
备注：Accepted at ICLR 2026 Conference

【9】De Novo Molecular Generation from Mass Spectra via Many-Body Enhanced Diffusion
标题：通过多体增强扩散从光谱中重新产生分子
链接：https://arxiv.org/abs/2602.01643

作者：Xichen Sun,Wentao Wei,Jiahua Rao,Jiancong Xie,Yuedong Yang

【10】Efficient Softmax Reformulation for Homomorphic Encryption via Moment Generating Function
标题：通过矩生成函数实现Homorphic加密的高效Softmax重构
链接：https://arxiv.org/abs/2602.01621

作者：Hanjun Park,Byeong-Seo Min,Jiheon Woo,Min-Wook Jeong,Jongho Shin,Yongwoo Lee,Young-Sik Kim,Yongjune Kim

【11】PromptRL: Prompt Matters in RL for Flow-Based Image Generation
标题：EntRL：RL中提示对于基于流的图像生成至关重要
链接：https://arxiv.org/abs/2602.01382

作者：Fu-Yun Wang,Han Zhang,Michael Gharbi,Hongsheng Li,Taesung Park

【12】Statistical MIA: Rethinking Membership Inference Attack for Reliable Unlearning Auditing
标题：统计MIA：重新思考会员推断攻击以实现可靠的无学习审计
链接：https://arxiv.org/abs/2602.01150

作者：Jialong Sun,Zeming Wei,Jiaxuan Zou,Jiacheng Gong,Guanheng Wang,Chengyang Dong,Jialong Li,Bo Liu

【13】Unifying Ranking and Generation in Query Auto-Completion via Retrieval-Augmented Generation and Multi-Objective Alignment
标题：通过检索增强生成和多目标对齐统一查询自动完成中的排名和生成
链接：https://arxiv.org/abs/2602.01023

作者：Kai Yuan,Anthony Zheng,Jia Hu,Divyanshu Sheth,Hemanth Velaga,Kylee Kim,Matteo Guarrera,Besim Avci,Xuetao Yin,Rajyashree Mukherjee,Sean Suchter
备注：11 pages, 4 figures

【14】RMFlow: Refined Mean Flow by a Noise-Injection Step for Multimodal Generation
标题：RMFlow：通过用于多峰生成的噪音注入步骤精制平均流量
链接：https://arxiv.org/abs/2602.00849

作者：Yuhao Huang,Shih-Hsin Wang,Andrea L. Bertozzi,Bao Wang
备注：Accepted to ICLR 2026

【15】Physics-informed Diffusion Generation for Geomagnetic Map Interpolation
标题：地磁图插值的物理信息扩散生成
链接：https://arxiv.org/abs/2602.00709

作者：Wenda Li,Tongya Zheng,Kaixuan Chen,Shunyu Liu,Haoze Jiang,Yunzhi Hao,Rui Miao,Zujie Ren,Mingli Song,Hang Shi,Gang Chen
备注：5 pages, 2 figures, IEEE ICASSP'26

【16】LocalV: Exploiting Information Locality for IP-level Verilog Generation
标题：LocalV：利用信息局部性实现IP级Verilog生成
链接：https://arxiv.org/abs/2602.00704

作者：Hanqi Lyu,Di Huang,Yaoyu Zhu,Kangcheng Liu,Bohan Dou,Chongxiao Li,Pengwei Jin,Shuyao Cheng,Rui Zhang,Zidong Du,Qi Guo,Xing Hu,Yunji Chen

【17】OD-DEAL: Dynamic Expert-Guided Adversarial Learning with Online Decomposition for Scalable Capacitated Vehicle Routing
标题：OD-DEAL：可扩展容量车辆路线的动态专家引导对抗学习和在线分解
链接：https://arxiv.org/abs/2602.00488

作者：Dongbin Jiao,Zisheng Chen,Xianyi Wang,Jintao Shi,Shengcai Liu,Shi Yan

【18】LatentTrack: Sequential Weight Generation via Latent Filtering
标题：LatentTrack：通过潜伏过滤生成序列权重
链接：https://arxiv.org/abs/2602.00458

作者：Omer Haq

【19】Open Materials Generation with Inference-Time Reinforcement Learning
标题：利用推理时强化学习生成开放材料
链接：https://arxiv.org/abs/2602.00424

作者：Philipp Hoellmer,Stefano Martiniani
备注：16 pages, 8 figures, 1 table

【20】Generation Order and Parallel Decoding in Masked Diffusion Models: An Information-Theoretic Perspective
标题：掩蔽扩散模型中的生成顺序和并行解码：信息论的视角
链接：https://arxiv.org/abs/2602.00286

作者：Shaorong Zhang,Longxuan Yu,Rob Brekelmans,Luhan Tang,Salman Asif,Greg Ver Steeg

【21】RPP: A Certified Poisoned-Sample Detection Framework for Backdoor Attacks under Dataset Imbalance
标题：RPP：数据集失衡下针对后门攻击的经过认证的中毒样本检测框架
链接：https://arxiv.org/abs/2602.00183

作者：Miao Lin,Feng Yu,Rui Ning,Lusi Li,Jiawei Chen,Qian Lou,Mengxin Zheng,Chunsheng Xin,Hongyi Wu

【22】The Illusion of Forgetting: Attack Unlearned Diffusion via Initial Latent Variable Optimization
标题：忘记的幻觉：通过初始潜在变量优化进行的攻击无习得扩散
链接：https://arxiv.org/abs/2602.00175

作者：Manyi Li,Yufan Liu,Lai Jiang,Bing Li,Yuming Li,Weiming Hu
备注：21 pages, 22 figures, 17 tables

【23】Learning Robust Reasoning through Guided Adversarial Self-Play
标题：通过引导对抗自我游戏学习稳健推理
链接：https://arxiv.org/abs/2602.00173

作者：Shuozhe Li,Vaishnav Tadiparthi,Kwonjoon Lee,Nakul Agarwal,Hossein Nourkhiz Mahjoub,Ehsan Moradi Pari,Lizhang Chen,Amy Zhang,Liu Leqi

【24】SPARC-RAG: Adaptive Sequential-Parallel Scaling with Context Management for Retrieval-Augmented Generation
标题：SPARC-RAG：具有上下文管理的自适应序列并行缩放，用于检索增强生成
链接：https://arxiv.org/abs/2602.00083

作者：Yuxin Yang,Gangda Deng,Ömer Faruk Akgül,Nima Chitsazan,Yash Govilkar,Akasha Tigalappanavara,Shi-Xiong Zhang,Sambit Sahu,Viktor Prasanna

【25】IntentCoding: Amplifying User Intent in Code Generation
标题：意图编码：在代码生成中放大用户意图
链接：https://arxiv.org/abs/2602.00066

作者：Zheng Fang,Yihong Dong,Lili Mou,Dongming Jin,Zhi Jin,Ge Li

【26】Generating Synthetic Health Sensor Data for Privacy-Preserving Wearable Stress Detection
标题：生成合成健康传感器数据以用于保护隐私的可穿戴压力检测
链接：https://arxiv.org/abs/2401.13327

作者：Lucas Lange,Nils Wenzlitschke,Erhard Rahm
备注：Published in the MDPI Sensors Journal

半/弱/无/有监督|不确定性|主动学习(24篇)

【1】MemSkill: Learning and Evolving Memory Skills for Self-Evolving Agents
标题：MemSkill：为自我进化的代理学习和进化记忆技能
链接：https://arxiv.org/abs/2602.02474

作者：Haozhen Zhang,Quanyu Long,Jianzhu Bao,Tao Feng,Weizhi Zhang,Haodong Yue,Wenya Wang
备注：Code is available at https://github.com/ViktorAxelsen/MemSkill

【2】Active Causal Experimentalist (ACE): Learning Intervention Strategies via Direct Preference Optimization
标题：主动因果实验主义者（ACE）：通过直接偏好优化的学习干预策略
链接：https://arxiv.org/abs/2602.02451

作者：Patrick Cooper,Alvaro Velasquez
备注：9 pages, 5 figures

【3】Active Transfer Bagging: A New Approach for Accelerated Active Learning Acquisition of Data by Combined Transfer Learning and Bagging Based Models
标题：主动转移打包：一种通过结合基于转移学习和打包的模型来加速主动学习数据获取的新方法
链接：https://arxiv.org/abs/2602.02415

作者：Vivienne Pelletier,Daniel J. Rivera,Obinna Nwokonkwo,Steven A. Wilson,Christopher L. Muhich

【4】Self-Supervised Learning from Structural Invariance
标题：结构不变性的自我监督学习
链接：https://arxiv.org/abs/2602.02381

作者：Yipeng Zhang,Hafez Ghaemi,Jungyoon Lee,Shahab Bakhtiari,Eilif B. Muller,Laurent Charlin
备注：ICLR 2026

【5】Unsupervised Physics-Informed Operator Learning through Multi-Stage Curriculum Training
标题：通过多阶段课程训练进行无监督的物理知情操作员学习
链接：https://arxiv.org/abs/2602.02264

作者：Paolo Marcandelli,Natansh Mathur,Stefano Markidis,Martina Siena,Stefano Mariani
备注：51 pages, 15 figures, 6 tables

【6】Learning While Staying Curious: Entropy-Preserving Supervised Fine-Tuning via Adaptive Self-Distillation for Large Reasoning Models
标题：保持好奇心的同时学习：通过大型推理模型的自适应自蒸馏进行保量监督微调
链接：https://arxiv.org/abs/2602.02244

作者：Hao Wang,Hao Gu,Hongming Piao,Kaixiong Gong,Yuxiao Ye,Xiangyu Yue,Sirui Han,Yike Guo,Dapeng Wu

【7】Active learning from positive and unlabeled examples
标题：从积极和未标记的例子中积极学习
链接：https://arxiv.org/abs/2602.02081

作者：Farnam Mansouri,Sandra Zilles,Shai Ben-David

【8】Observation-dependent Bayesian active learning via input-warped Gaussian processes
标题：通过输入扭曲高斯过程的观察相关Bayesian主动学习
链接：https://arxiv.org/abs/2602.01898

作者：Sanna Jarl,Maria Bånkestad,Jonathan J. S. Scragg,Jens Sjölund
备注：13 pages

【9】Quantifying Epistemic Predictive Uncertainty in Conformal Prediction
标题：量化保形预测中的认识预测不确定性
链接：https://arxiv.org/abs/2602.01667

作者：Siu Lun Chau,Soroush H. Zargarbashi,Yusuf Sale,Michele Caprio
备注：42 pages

【10】AdaptNC: Adaptive Nonconformity Scores for Uncertainty-Aware Autonomous Systems in Dynamic Environments
标题：AdaptNC：动态环境中具有不确定性的自治系统的自适应不一致性分数
链接：https://arxiv.org/abs/2602.01629

作者：Renukanandan Tumu,Aditya Singh,Rahul Mangharam

【11】SUSD: Structured Unsupervised Skill Discovery through State Factorization
标题：SUUSD：通过状态分解实现结构化无监督技能发现
链接：https://arxiv.org/abs/2602.01619

作者：Seyed Mohammad Hadi Hosseini,Mahdieh Soleymani Baghshah
备注：Accepted as a conference paper at ICLR 2026

【12】Sample Efficient Active Algorithms for Offline Reinforcement Learning
标题：离线强化学习的高效主动算法示例
链接：https://arxiv.org/abs/2602.01260

作者：Soumyadeep Roy,Shashwat Kushwaha,Ambedkar Dukkipati

【13】Supervised Fine-Tuning Needs to Unlock the Potential of Token Priority
标题：监督微调需要充分发挥代币优先级的潜力
链接：https://arxiv.org/abs/2602.01227

作者：Zhanming Shen,Zeyu Qin,Jiaqi Hu,Wentao Ye,Hao Chen,Xiaomeng Hu,Haokai Xu,Gang Chen,Yi R. Fung,Haobo Wang

【14】Multi-Fidelity Physics-Informed Neural Networks with Bayesian Uncertainty Quantification and Adaptive Residual Learning for Efficient Solution of Parametric Partial Differential Equations
标题：具有Bayesian不确定性量化和自适应剩余学习的多保真物理信息神经网络用于有效求解参数偏微方程
链接：https://arxiv.org/abs/2602.01176

作者：Olaf Yunus Laitinen Imanov
备注：8 pages, 4 figures, 6 tables

【15】Forest-Guided Semantic Transport for Label-Supervised Manifold Alignment
标题：用于标签监督的Manifold对齐的森林引导语义传输
链接：https://arxiv.org/abs/2602.00974

作者：Adrien Aumon,Myriam Lizotte,Guy Wolf,Kevin R. Moon,Jake S. Rhodes

【16】Beyond What Seems Necessary: Hidden Gains from Scaling Training-Time Reasoning Length under Outcome Supervision
标题：超出看似必要的范围：在结果监督下扩大训练时间推理长度的隐藏收益
链接：https://arxiv.org/abs/2602.00927

作者：Yihao Xue,Allan Zhang,Jianhao Huang,Amit Sahai,Baharan Mirzasoleiman

【17】Learning Heat-based Equations in Self-similar variables
标题：学习自相似变量中的热基方程
链接：https://arxiv.org/abs/2602.00872

作者：Shihao Wang,Qipeng Qian,Jingquan Wang

【18】Forget by Uncertainty: Orthogonal Entropy Unlearning for Quantized Neural Networks
标题：被不确定性遗忘：量化神经网络的正交信息去学习
链接：https://arxiv.org/abs/2602.00567

作者：Tian Zhang,Yujia Tong,Junhao Dong,Ke Xu,Yuze Wang,Jingling Yuan

【19】Beyond the Loss Curve: Scaling Laws, Active Learning, and the Limits of Learning from Exact Posteriors
标题：超越损失曲线：缩放定律、主动学习以及从精确事后学习的局限性
链接：https://arxiv.org/abs/2602.00315

作者：Arian Khorasani,Nathaniel Chen,Yug D Oswal,Akshat Santhana Gopalan,Egemen Kolemen,Ravid Shwartz-Ziv

【20】Interpretable Unsupervised Deformable Image Registration via Confidence-bound Multi-Hop Visual Reasoning
标题：通过保密多跳视觉推理的可解释无监督可变形图像配准
链接：https://arxiv.org/abs/2602.00211

作者：Zafar Iqbal,Anwar Ul Haq,Srimannarayana Grandhi

【21】How Understanding Forecast Uncertainty Resolves the Explainability Problem in Machine Learning Models
标题：了解预测不确定性如何解决机器学习模型中的可解释性问题
链接：https://arxiv.org/abs/2602.00179

作者：Joseph L. Breeden
备注：22 pages; 2 figures

【22】The Impact of Machine Learning Uncertainty on the Robustness of Counterfactual Explanations
标题：机器学习不确定性对反事实解释稳健性的影响
链接：https://arxiv.org/abs/2602.00063

作者：Leonidas Christodoulou,Chang Sun

【23】Propagating the prior from far to near offset: A self-supervised diffusion framework for progressively recovering near-offsets of towed-streamer data
标题：从远距到近距叠加先验：用于逐步恢复拖曳拖缆数据近距叠加的自我监督扩散框架
链接：https://arxiv.org/abs/2602.01909

作者：Shijun Cheng,Tariq Alkhalifah

【24】Uncertainty-Aware Multimodal Learning via Conformal Shapley Intervals
标题：通过保形Shapley区间的不确定性感知多模式学习
链接：https://arxiv.org/abs/2602.00171

作者：Mathew Chandy,Michael Johnson,Judong Shen,Devan V. Mehrotra,Hua Zhou,Jin Zhou,Xiaowu Dai

迁移|Zero/Few/One-Shot|自适应(34篇)

【1】Energy-Efficient Neuromorphic Computing for Edge AI: A Framework with Adaptive Spiking Neural Networks and Hardware-Aware Optimization
标题：边缘人工智能的节能神经形态计算：具有自适应尖峰神经网络和硬件感知优化的框架
链接：https://arxiv.org/abs/2602.02439

作者：Olaf Yunus Laitinen Imanov,Derya Umut Kulali,Taner Yilmaz,Duygu Erisken,Rana Irem Turhan
备注：8 pages, 4 figures, 4 tables. Submitted to IEEE Transactions on Neural Networks and Learning Systems (TNNLS)

【2】NAB: Neural Adaptive Binning for Sparse-View CT reconstruction
标题：NAB：用于稀疏视图CT重建的神经自适应分组
链接：https://arxiv.org/abs/2602.02356

作者：Wangduo Xie,Matthew B. Blaschko

【3】Alignment-Aware Model Adaptation via Feedback-Guided Optimization
标题：通过反馈引导优化实现对齐感知模型自适应
链接：https://arxiv.org/abs/2602.02258

作者：Gaurav Bhatt,Aditya Chinchure,Jiawei Zhou,Leonid Sigal

【4】DCoPilot: Generative AI-Empowered Policy Adaptation for Dynamic Data Center Operations
标题：DCoPilot：用于动态数据中心运营的生成式AI授权策略调整
链接：https://arxiv.org/abs/2602.02137

作者：Minghao Li,Ruihang Wang,Rui Tan,Yonggang Wen

【5】Calibrating Adaptive Smoothing Methods for Freeway Traffic Reconstruction
标题：高速公路交通重构中自适应平滑方法的标定
链接：https://arxiv.org/abs/2602.02072

作者：Junyi Ji,Derek Gloudemans,Gergely Zachár,Matthew Nice,William Barbour,Daniel B. Work

【6】Adaptive Quality-Diversity Trade-offs for Large-Scale Batch Recommendation
标题：大规模批量推荐的自适应质量多样性权衡
链接：https://arxiv.org/abs/2602.02024

作者：Clémence Réda,Tomas Rigaux,Hiba Bederina,Koh Takeuchi,Hisashi Kashima,Jill-Jênn Vie

【7】Zero-Shot Off-Policy Learning
标题：零攻击政策外学习
链接：https://arxiv.org/abs/2602.01962

作者：Arip Asadulaev,Maksim Bobrin,Salem Lahlou,Dmitry Dylov,Fakhri Karray,Martin Takac

【8】PIMCST: Physics-Informed Multi-Phase Consensus and Spatio-Temporal Few-Shot Learning for Traffic Flow Forecasting
标题：PIMCST：用于交通流预测的物理知情多阶段共识和时空Few-Shot学习
链接：https://arxiv.org/abs/2602.01936

作者：Abdul Joseph Fofanah,Lian Wen,David Chen

【9】Efficient Cross-Architecture Knowledge Transfer for Large-Scale Online User Response Prediction
标题：用于大规模在线用户响应预测的高效跨架构知识转移
链接：https://arxiv.org/abs/2602.01775

作者：Yucheng Wu,Yuekui Yang,Hongzheng Li,Anan Liu,Jian Xiao,Junjie Zhai,Huan Yu,Shaoping Ma,Leye Wang
备注：15 pages

【10】DIA-CLIP: a universal representation learning framework for zero-shot DIA proteomics
标题：DIA-CLIP：一个通用的Zero-ShotDIA蛋白质组学表征学习框架
链接：https://arxiv.org/abs/2602.01772

作者：Yucheng Liao,Han Wen,Weinan E,Weijie Zhang
备注：21 pages, 5 figures

【11】Probability-Entropy Calibration: An Elastic Indicator for Adaptive Fine-tuning
标题：概率-熵校准：自适应微调的弹性指标
链接：https://arxiv.org/abs/2602.01745

作者：Wenhao Yu,Shaohang Wei,Jiahong Liu,Yifan Li,Minda Hu,Aiwei Liu,Hao Zhang,Irwin King

【12】ASGMamba: Adaptive Spectral Gating Mamba for Multivariate Time Series Forecasting
标题：ASGMamba：用于多元时间序列预测的自适应光谱门控Mamba
链接：https://arxiv.org/abs/2602.01668

作者：Qianyang Li,Xingjun Zhang,Shaoxun Wang,Jia Wei,Yueqi Xing

【13】COMET: Codebook-based Online-adaptive Multi-scale Embedding for Time-series Anomaly Detection
标题：COMET：基于码本的在线自适应多尺度嵌入时间序列异常检测
链接：https://arxiv.org/abs/2602.01635

作者：Jinwoo Park,Hyeongwon Kang,Seung Hun Han,Pilsung Kang

【14】Adaptive Rollout Allocation for Online Reinforcement Learning with Verifiable Rewards
标题：具有可验证奖励的在线强化学习自适应推出分配
链接：https://arxiv.org/abs/2602.01601

作者：Hieu Trung Nguyen,Bao Nguyen,Wenao Ma,Yuzhi Zhao,Ruifeng She,Viet Anh Nguyen
备注：Accepted at ICLR 2026

【15】White-Box Neural Ensemble for Vehicular Plasticity: Quantifying the Efficiency Cost of Symbolic Auditability in Adaptive NMPC
标题：车辆可塑性的白盒神经集合：量化自适应NMPC中符号可审计性的效率成本
链接：https://arxiv.org/abs/2602.01516

作者：Enzo Nicolas Spotorno,Matheus Wagner,Antonio Augusto Medeiros Frohlich
备注：5 pages, 1 table, 1 figure, submitted to IEEE VTC 2026 Recent Results Track

【16】The Gaussian-Head OFL Family: One-Shot Federated Learning from Client Global Statistics
标题：Gaussian-Head OFL家族：来自客户全球统计的一站式联邦学习
链接：https://arxiv.org/abs/2602.01186

作者：Fabio Turazza,Marco Picone,Marco Mamei
备注：Accepted at the International Conference on Learning Representations (ICLR) 2026

【17】WinFLoRA: Incentivizing Client-Adaptive Aggregation in Federated LoRA under Privacy Heterogeneity
标题：WinFLoRA：在隐私异类下激励联邦LoRA中的客户端自适应聚合
链接：https://arxiv.org/abs/2602.01126

作者：Mengsha Kou,Xiaoyu Xia,Ziqi Wang,Ibrahim Khalil,Runkun Luo,Jingwen Zhou,Minhui Xue
备注：12 pages

【18】SwiftRepertoire: Few-Shot Immune-Signature Synthesis via Dynamic Kernel Codes
标题：SwiftRepertoire：基于动态核代码的少次免疫签名合成
链接：https://arxiv.org/abs/2602.01051

作者：Rong Fu,Wenxin Zhang,Muge Qi,Yang Li,Yabin Jin,Jiekai Wu,Jiaxuan Lu,Chunlei Meng,Youjin Wang,Zeli Su,Juntao Gao,Li Bao,Qi Zhao,Wei Luo,Simon Fong
备注：19 pages, 8 figures, 8 tables

【19】Adaptive Dual-Weighting Framework for Federated Learning via Out-of-Distribution Detection
标题：通过分布外检测的联邦学习自适应双加权框架
链接：https://arxiv.org/abs/2602.01039

作者：Zhiwei Ling,Hailiang Zhao,Chao Zhang,Xiang Ao,Ziqi Wang,Cheng Zhang,Zhen Qin,Xinkui Zhao,Kingsum Chow,Yuanqing Wu,MengChu Zhou

【20】From drift to adaptation to the failed ml model: Transfer Learning in Industrial MLOps
标题：从漂移到适应失败的ml模型：工业MLOps中的迁移学习
链接：https://arxiv.org/abs/2602.00957

作者：Waqar Muhammad Ashraf,Talha Ansar,Fahad Ahmed,Jawad Hussain,Muhammad Mujtaba Abbas,Vivek Dua
备注：Corresponding author: v.dua@ucl.ac.uk

【21】SALAAD: Sparse And Low-Rank Adaptation via ADMM
标题：SALAAD：通过ADMM进行稀疏和低等级适应
链接：https://arxiv.org/abs/2602.00942

作者：Hao Ma,Melis Ilayda Bal,Liang Zhang,Bingcong Li,Niao He,Melanie Zeilinger,Michael Muehlebach

【22】Domain-Adaptive and Scalable Dense Retrieval for Content-Based Recommendation
标题：面向内容推荐的领域自适应可扩展密集检索
链接：https://arxiv.org/abs/2602.00899

作者：Mritunjay Pandey
备注：13 pages, 4 figures. Semantic dense retrieval for content-based recommendation on Amazon Reviews 2023 (Category - Fashion). Dataset statistics: 2.0M users; 825.9K items; 2.5M ratings; 94.9M review tokens; 510.5M metadata tokens. Timespan: May 1996 to September 2023. Metadata includes: user reviews (ratings, text, helpfulness votes, etc.); item metadata (descriptions, price, raw images, etc.)

【23】Spectral Imbalance Causes Forgetting in Low-Rank Continual Adaptation
标题：光谱不平衡导致低阶连续适应中的遗忘
链接：https://arxiv.org/abs/2602.00722

作者：Hao Gu,Mao-Lin Luo,Zi-Hao Zhou,Han-Chen Zhang,Min-Ling Zhang,Tong Wei
备注：19 pages, 6 figures

【24】Rethinking Zero-Shot Time Series Classification: From Task-specific Classifiers to In-Context Inference
标题：重新思考Zero-Shot时间序列分类：从特定任务的分类器到上下文推理
链接：https://arxiv.org/abs/2602.00620

作者：Juntao Fang,Shifeng Xie,Shengbin Nie,Yuhui Ling,Yuming Liu,Zijian Li,Keli Zhang,Lujia Pan,Themis Palpanas,Ruichu Cai

【25】NPNet: A Non-Parametric Network with Adaptive Gaussian-Fourier Positional Encoding for 3D Classification and Segmentation
标题：NPNet：一个具有自适应高斯-傅里叶位置编码的非参数网络，用于3D分类和分割
链接：https://arxiv.org/abs/2602.00542

作者：Mohammad Saeid,Amir Salarpour,Pedram MohajerAnsari,Mert D. Pesé
备注：Accepted to the 2026 IEEE Intelligent Vehicles Symposium (IV 2026)

【26】Adaptive Momentum and Nonlinear Damping for Neural Network Training
标题：神经网络训练的自适应动量和非线性衰减
链接：https://arxiv.org/abs/2602.00334

作者：Aikaterini Karoni,Rajit Rajpal,Benedict Leimkuhler,Gabriel Stoltz
备注：29 pages, 11 figures

【27】RAPTOR: Ridge-Adaptive Logistic Probes
标题：报道者：山脊自适应逻辑探针
链接：https://arxiv.org/abs/2602.00158

作者：Ziqi Gao,Yaotian Zhu,Qingcheng Zeng,Xu Zhao,Ziqing Wang,Feng Ruan,Kaize Ding
备注：Preprint

【28】1S-DAug: One-Shot Data Augmentation for Robust Few-Shot Generalization
标题：1 S-DAug：单次数据增强，实现稳健的少次概括
链接：https://arxiv.org/abs/2602.00114

作者：Yunwei Bai,Ying Kiat Tan,Yao Shu,Tsuhan Chen

【29】Generative AI-enhanced Probabilistic Multi-Fidelity Surrogate Modeling Via Transfer Learning
标题：通过迁移学习的生成式AI增强的概率多保真度代理建模
链接：https://arxiv.org/abs/2602.00072

作者：Jice Zeng,David Barajas-Solano,Hui Chen

【30】Extending Beacon to Hindi: Cultural Adaptation Drives Cross-Lingual Sycophancy
标题：将Beacon扩展到印地语：文化适应推动跨语言谄媚
链接：https://arxiv.org/abs/2602.00046

作者：Sarthak Sattigeri
备注：First Hindi sycophancy benchmark using a three-condition design separating language and cultural effects, with empirical evaluation across four instruction-tuned models

【31】Culinary Crossroads: A RAG Framework for Enhancing Diversity in Cross-Cultural Recipe Adaptation
标题：烹饪十字路口：增强跨文化食谱适应多样性的RAG框架
链接：https://arxiv.org/abs/2507.21934

作者：Tianyi Hu,Andrea Morales-Garzón,Jingyi Zheng,Maria Maistro,Daniel Hershcovich

【32】Transfer Learning Through Conditional Quantile Matching
标题：通过条件分位数匹配的迁移学习
链接：https://arxiv.org/abs/2602.02358

作者：Yikun Zhang,Steven Wilkins-Reeves,Wesley Lee,Aude Hofleitner
备注：24 pages (8 pages for the main paper), 3 figures, 3 tables

【33】Robust Generalization with Adaptive Optimal Transport Priors for Decision-Focused Learning
标题：具有自适应最佳传输先验的鲁棒概括，用于以决策为中心的学习
链接：https://arxiv.org/abs/2602.01427

作者：Haixiang Sun,Andrew L. Liu

【34】Test-Time Adaptation for Non-stationary Time Series: From Synthetic Regime Shifts to Financial Markets
标题：非平稳时间序列的测试时间适应：从合成制度转变到金融市场
链接：https://arxiv.org/abs/2602.00073

作者：Yurui Wu,Qingying Deng,Wonou Chung,Mairui Li

强化学习(25篇)

【1】Expanding the Capabilities of Reinforcement Learning via Text Feedback
标题：通过文本反馈扩展强化学习的能力
链接：https://arxiv.org/abs/2602.02482

作者：Yuda Song,Lili Chen,Fahim Tajwar,Remi Munos,Deepak Pathak,J. Andrew Bagnell,Aarti Singh,Andrea Zanette
备注：43 pages, 6 figures

【2】David vs. Goliath: Verifiable Agent-to-Agent Jailbreaking via Reinforcement Learning
标题：大卫与歌利亚：通过强化学习进行可验证的代理人对代理人越狱
链接：https://arxiv.org/abs/2602.02395

作者：Samuel Nellessen,Tal Kachman
备注：Under review. 8 main pages, 2 figures, 2 tables. Appendix included

【3】ECHO-2: A Large Scale Distributed Rollout Framework for Cost-efficient Reinforcement Learning
标题：ECHO-2：一个用于经济高效强化学习的大规模分布式推出框架
链接：https://arxiv.org/abs/2602.02192

作者：Jie Xiao,Meng Chen,Qingnan Ren,Song Jingwei,Jiaqi Huang,Yangshen Deng,Chris Tong,Wanyi Chen,Suli Wang,Ziqian Bi,Shuo Lu,Yiqun Duan,Lynn Ai,Eric Yang,Bill Shi
备注：23 pages, 7 figures

【4】ECHO: Entropy-Confidence Hybrid Optimization for Test-Time Reinforcement Learning
标题：ECHO：测试时强化学习的熵-置信度混合优化
链接：https://arxiv.org/abs/2602.02150

作者：Chu Zhao,Enneng Yang,Yuting Liu,Jianzhe Zhao,Guibing Guo
备注：19 ppages

【5】Probabilistic Performance Guarantees for Multi-Task Reinforcement Learning
标题：多任务强化学习的概率性能保证
链接：https://arxiv.org/abs/2602.02098

作者：Yannik Schnitzer,Mathias Jackermeier,Alessandro Abate,David Parker

【6】FORLER: Federated Offline Reinforcement Learning with Q-Ensemble and Actor Rectification
标题：FORLER：具有Q-Ensemble和Actor Rectification的联合离线强化学习
链接：https://arxiv.org/abs/2602.02055

作者：Nan Qiao,Sheng Yue
备注：accetped by IEEE International Conference on Communications (ICC 2026)

【7】Beyond Mode Elicitation: Diversity-Preserving Reinforcement Learning via Latent Diffusion Reasoner
标题：超越模式激发：通过潜在扩散推理的多样性保持强化学习
链接：https://arxiv.org/abs/2602.01705

作者：Haoqiang Kang,Yizhe Zhang,Nikki Lijing Kuang,Yi-An Ma,Lianhui Qin

【8】TABX: A High-Throughput Sandbox Battle Simulator for Multi-Agent Reinforcement Learning
标题：TAGX：用于多智能体强化学习的高吞吐量沙盒战斗模拟器
链接：https://arxiv.org/abs/2602.01665

作者：Hayeong Lee,JunHyeok Oh,Byung-Jun Lee

【9】FlowSteer: Interactive Agentic Workflow Orchestration via End-to-End Reinforcement Learning
标题：FlowSteer：通过端到端强化学习进行交互式统计工作流程规划
链接：https://arxiv.org/abs/2602.01664

作者：Mingda Zhang,Haoran Luo,Tiesunlong Shen,Qika Lin,Xiaoying Tang,Rui Mao,Erik Cambria
备注：41 pages, 7 figures, 6 tables. Project page: http://flowsteer.org/

【10】Boosting Maximum Entropy Reinforcement Learning via One-Step Flow Matching
标题：通过一步流匹配提升最大熵强化学习
链接：https://arxiv.org/abs/2602.01606

作者：Zeqiao Li,Yijing Wang,Haoyu Wang,Zheng Li,Zhiqiang Zuo

【11】The Enhanced Physics-Informed Kolmogorov-Arnold Networks: Applications of Newton's Laws in Financial Deep Reinforcement Learning (RL) Algorithms
标题：增强的物理信息Kolmogorov-Arnold网络：牛顿定律在金融深度强化学习（RL）算法中的应用
链接：https://arxiv.org/abs/2602.01388

作者：Trang Thoi,Hung Tran,Tram Thoi,Huaiyang Zhong

【12】When Domains Interact: Asymmetric and Order-Sensitive Cross-Domain Effects in Reinforcement Learning for Reasoning
标题：当领域相互作用时：推理强化学习中的不对称和顺序敏感跨领域效应
链接：https://arxiv.org/abs/2602.01365

作者：Wang Yang,Shouren Wang,Chaoda Song,Chuang Ma,Xinpeng Li,Nengbo Wang,Kaixiong Zhou,Vipin Chaudhary,Xiaotian Han

【13】CRAFT: Calibrated Reasoning with Answer-Faithful Traces via Reinforcement Learning for Multi-Hop Question Answering
标题：CRAFT：通过多跳问题回答的强化学习，用忠诚痕迹进行校准推理
链接：https://arxiv.org/abs/2602.01348

作者：Yu Liu,Wenxiao Zhang,Cong Cao,Fangfang Yuan,Weizhuo Chen,Cheng Hu,Pin Xu,Yuling Yang,Kun Peng,Diandian Guo,Qiang Sun,Yanbing Liu,Jin B. Hong,Zhiyuan Ma

【14】Mixture-of-World Models: Scaling Multi-Task Reinforcement Learning with Modular Latent Dynamics
标题：混合世界模型：利用模块化潜在动力学扩展多任务强化学习
链接：https://arxiv.org/abs/2602.01270

作者：Boxuan Zhang,Weipu Zhang,Zhaohan Feng,Wei Xiao,Jian Sun,Jie Chen,Gang Wang

【15】PolicyFlow: Policy Optimization with Continuous Normalizing Flow in Reinforcement Learning
标题：Policy Flow：强化学习中通过持续规范化流程进行政策优化
链接：https://arxiv.org/abs/2602.01156

作者：Shunpeng Yang,Ben Liu,Hua Chen
备注：Submitted to ICLR 2026

【16】Good SFT Optimizes for SFT, Better SFT Prepares for Reinforcement Learning
标题：良好的SFT优化SFT，更好的SFT为强化学习做好准备
链接：https://arxiv.org/abs/2602.01058

作者：Dylan Zhang,Yufeng Xu,Haojin Wang,Qingzhi Chen,Hao Peng

【17】Communications-Incentivized Collaborative Reasoning in NetGPT through Agentic Reinforcement Learning
标题：通过抽象强化学习在NetGPT中激励通信协作推理
链接：https://arxiv.org/abs/2602.00766

作者：Xiaoxue Yu,Rongpeng Li,Zhifeng Zhao,Honggang Zhang

【18】Reinforcement Learning-assisted Constraint Relaxation for Constrained Expensive Optimization
标题：约束昂贵优化的强化学习辅助约束松弛
链接：https://arxiv.org/abs/2602.00532

作者：Qianhao Zhu,Sijie Ma,Zeyuan Ma,Hongshu Guo,Yue-Jiao Gong

【19】Search Inspired Exploration in Reinforcement Learning
标题：强化学习中的搜索启发式探索
链接：https://arxiv.org/abs/2602.00460

作者：Georgios Sotirchos,Zlatan Ajanović,Jens Kober

【20】Sample Complexity Analysis for Constrained Bilevel Reinforcement Learning
标题：约束二层强化学习的样本复杂性分析
链接：https://arxiv.org/abs/2602.00282

作者：Naman Saxena,Vaneet Aggarwal

【21】Distributional Reinforcement Learning for Condition-Based Maintenance of Multi-Pump Equipment
标题：多泵设备状态维护的分布式强化学习
链接：https://arxiv.org/abs/2602.00051

作者：Takato Yasuno
备注：15 pages, 15 figures

【22】Asynchronous MultiAgent Reinforcement Learning for 5G Routing under Side Constraints
标题：侧约束下5G路由的同步多Agent强化学习
链接：https://arxiv.org/abs/2602.00035

作者：Sebastian Racedo,Brigitte Jaumard,Oscar Delgado,Meysam Masoudi

【23】Representation Learning Enhanced Deep Reinforcement Learning for Optimal Operation of Hydrogen-based Multi-Energy Systems
标题：表示学习增强的深度强化学习用于氢基多能源系统的优化运行
链接：https://arxiv.org/abs/2602.00027

作者：Zhenyu Pu,Yu Yang,Lun Yang,Qing-Shan Jia,Xiaohong Guan,Costas J. Spanos
备注：14 pages, 7 figures

【24】Reinforcement Learning via Conservative Agent for Environments with Random Delays
标题：随机延迟环境中的保守代理强化学习
链接：https://arxiv.org/abs/2507.18992

作者：Jongsoo Lee,Jangwon Kim,Jiseok Jeong,Soohee Han

【25】Reinforcement Learning for Control Systems with Time Delays: A Comprehensive Survey
标题：具有时间延迟的控制系统的强化学习：全面调查
链接：https://arxiv.org/abs/2602.00399

作者：Armando Alves Neto
备注：30 pages

元学习(1篇)

【1】Trust Region Continual Learning as an Implicit Meta-Learner
标题：作为内隐元学习者的信任区持续学习
链接：https://arxiv.org/abs/2602.02417

作者：Zekun Wang,Anant Gupta,Christopher J. MacLellan
备注：19 pages, 23 tables

符号|符号学习(2篇)

【1】Enhancing Generalization in Evolutionary Feature Construction for Symbolic Regression through Vicinal Jensen Gap Minimization
标题：通过邻近Jensen差距最小化增强符号回归进化特征构建中的推广
链接：https://arxiv.org/abs/2602.01510

作者：Hengzhe Zhang,Qi Chen,Bing Xue,Wolfgang Banzhaf,Mengjie Zhang

【2】From Numbers to Prompts: A Cognitive Symbolic Transition Mechanism for Lightweight Time-Series Forecasting
标题：从数字到预测：轻量级时间序列预测的认知符号转换机制
链接：https://arxiv.org/abs/2602.00088

作者：Namkyung Yoon,Hwangnam Kim
备注：16 pages, 5 figures. Submitted to ACM Transactions on Intelligent Systems and Technology

分层学习(1篇)

【1】Hierarchical Federated Learning with SignSGD: A Highly Communication-Efficient Approach
标题：使用SignBCD的分层联邦学习：一种高沟通效率的方法
链接：https://arxiv.org/abs/2602.02355

作者：Amirreza Kazemi,Seyed Mohammad Azimi-Abarghouyi,Gabor Fodor,Carlo Fischione

医学相关(12篇)

【1】Advancing General-Purpose Reasoning Models with Modular Gradient Surgery
标题：通过模块化梯度手术推进通用推理模型
链接：https://arxiv.org/abs/2602.02301

作者：Min Cai,Yu Liang,Longzheng Wang,Yan Wang,Yueyang Zhang,Long Xia,Zhiyuan Sun,Xi Ye,Daiting Shi
备注：Preprint; Code: https://github.com/StringNLPLAB/MGS; Website: https://modular-gradient-surgery.github.io

【2】MoLF: Mixture-of-Latent-Flow for Pan-Cancer Spatial Gene Expression Prediction from Histology
标题：MoLF：从组织学预测泛癌症空间基因表达的混合潜伏流
链接：https://arxiv.org/abs/2602.02282

作者：Susu Hu,Stefanie Speidel

【3】Toxicity Assessment in Preclinical Histopathology via Class-Aware Mahalanobis Distance for Known and Novel Anomalies
标题：通过已知和新异常的已知Mahalanobis距离在临床前组织病理学中进行毒性评估
链接：https://arxiv.org/abs/2602.02124

作者：Olga Graf,Dhrupal Patel,Peter Groß,Charlotte Lempp,Matthias Hein,Fabian Heinemann

【4】Bayesian Integration of Nonlinear Incomplete Clinical Data
标题：非线性不完整临床数据的Bayesian积分
链接：https://arxiv.org/abs/2602.01924

作者：Lucía González-Zamorano,Nuria Balbás-Esteban,Vanessa Gómez-Verdejo,Albert Belenguer-Llorens,Carlos Sevilla-Salcedo

【5】CortiNet: A Physics-Perception Hybrid Cortical-Inspired Dual-Stream Network for Gallbladder Disease Diagnosis from Ultrasound
标题：CortiNet：一种物理感知混合皮质启发的双流网络，用于超声诊断胆囊疾病
链接：https://arxiv.org/abs/2602.01000

作者：Vagish Kumar,Souvik Chakraborty

【6】Hybrid Topological and Deep Feature Fusion for Accurate MRI-Based Alzheimer's Disease Severity Classification
标题：混合布局和深度特征融合实现基于MRI的阿尔茨海默病严重程度分类
链接：https://arxiv.org/abs/2602.00956

作者：Faisal Ahmed
备注：20 pages, 6 Figures

【7】Efficient Deep Learning for Medical Imaging: Bridging the Gap Between High-Performance AI and Clinical Deployment
标题：医学成像的高效深度学习：弥合高性能人工智能和临床部署之间的差距
链接：https://arxiv.org/abs/2602.00910

作者：Cuong Manh Nguyen,Truong-Son Hy

【8】Sheaf Neural Networks and biomedical applications
标题：Sheaf神经网络和生物医学应用
链接：https://arxiv.org/abs/2602.00159

作者：Aneeqa Mehrab,Jan Willem Van Looy,Pietro Demurtas,Stefano Iotti,Emil Malucelli,Francesca Rossi,Ferdinando Zanchetta,Rita Fioresi

【9】Quantum Model Parallelism for MRI-Based Classification of Alzheimer's Disease Stages
标题：基于MRI的阿尔茨海默病分期分类的量子模型平行论
链接：https://arxiv.org/abs/2602.00128

作者：Emine Akpinar,Murat Oduncuoglu
备注：Under review at Quantum Machine Intelligence (Springer Nature)

【10】AI-assisted Protocol Information Extraction For Improved Accuracy and Efficiency in Clinical Trial Workflows
标题：AI辅助协议信息提取以提高临床试验工作流程的准确性和效率
链接：https://arxiv.org/abs/2602.00052

作者：Ramtin Babaeipour,François Charest,Madison Wright

【11】Privacy in Practice: Private COVID-19 Detection in X-Ray Images (Extended Version)
标题：实践中的隐私：X射线图像中的私人COVID-19检测（扩展版本）
链接：https://arxiv.org/abs/2211.11434

作者：Lucas Lange,Maja Schneider,Peter Christen,Erhard Rahm
备注：Extended version of the paper accepted at the 20th International Conference on Security and Cryptography SECRYPT 2023. This version is more detailed and includes additional content: a longer results chapter and an appendix containing a proof

【12】Radiomics in Medical Imaging: Methods, Applications, and Challenges
标题：医学成像中的放射组学：方法、应用和挑战
链接：https://arxiv.org/abs/2602.00102

作者：Fnu Neha,Deepak kumar Shukla

蒸馏|知识提取(6篇)

【1】BicKD: Bilateral Contrastive Knowledge Distillation
标题：BicKD：双边对比知识提炼
链接：https://arxiv.org/abs/2602.01265

作者：Jiangnan Zhu,Yukai Xu,Li Xiong,Yixuan Liu,Junxu Liu,Hong kyu Lee,Yujie Gu

【2】Probing the Knowledge Boundary: An Interactive Agentic Framework for Deep Knowledge Extraction
标题：探索知识边界：深度知识提取的交互式统计框架
链接：https://arxiv.org/abs/2602.00959

作者：Yuheng Yang,Siqi Zhu,Tao Feng,Ge Liu,Jiaxuan You
备注：Homepage: https://ulab-uiuc.github.io/KnowledgeExtraction/

【3】Investigating the Robustness of Subtask Distillation under Spurious Correlation
标题：伪相关下子任务蒸馏的鲁棒性研究
链接：https://arxiv.org/abs/2602.00852

作者：Pattarawat Chormai,Klaus-Robert Müller,Grégoire Montavon
备注：7 pages, 3 figures

【4】Audio-to-Image Bird Species Retrieval without Audio-Image Pairs via Text Distillation
标题：通过文本蒸馏进行无音频图像对的音频到图像鸟类物种检索
链接：https://arxiv.org/abs/2602.00681

作者：Ilyass Moummad,Marius Miron,Lukas Rauch,David Robinson,Alexis Joly,Olivier Pietquin,Emmanuel Chemla,Matthieu Geist

【5】CoRe-Fed: Bridging Collaborative and Representation Fairness via Federated Embedding Distillation
标题：核心反馈：通过联邦嵌入蒸馏来桥梁协作和代表公平
链接：https://arxiv.org/abs/2602.00647

作者：Noorain Mukhtiar,Adnan Mahmood,Quan Z. Sheng
备注：7 pages (main content), 2 pages (references), Accepted in AAAI 2026

【6】Post-Training Probability Manifold Correction via Structured SVD Pruning and Self-Referential Distillation
标题：通过结构化MVD修剪和自参考蒸馏进行训练后概率Manifest纠正
链接：https://arxiv.org/abs/2602.00372

作者：Aaron R. Flouro,Shawn P. Chadwick
备注：16 pages, 10 tables, 4 figures

推荐(1篇)

【1】Dynamic Prior Thompson Sampling for Cold-Start Exploration in Recommender Systems
标题：推荐系统中冷启动探索的动态先验Thompson抽样
链接：https://arxiv.org/abs/2602.00943

作者：Zhenyu Zhao,David Zhang,Ellie Zhao,Ehsan Saberian

聚类(1篇)

【1】Event Driven Clustering Algorithm
标题：事件驱动的集群算法
链接：https://arxiv.org/abs/2602.00115

作者：David El-Chai Ben-Ezra,Adar Tal,Daniel Brisk
备注：~10 pages, 2 figures

超分辨率|去噪|去模糊|去雾(1篇)

【1】Geometry- and Relation-Aware Diffusion for EEG Super-Resolution
标题：脑电超分辨率的几何和几何感知扩散
链接：https://arxiv.org/abs/2602.02238

作者：Laura Yao,Gengwei Zhang,Moajjem Chowdhury,Yunmei Liu,Tianlong Chen

自动驾驶|车辆|车道检测等(1篇)

【1】Online Fine-Tuning of Pretrained Controllers for Autonomous Driving via Real-Time Recurrent RL
标题：通过实时回归RL对自动驾驶预训练控制器进行在线微调
链接：https://arxiv.org/abs/2602.02236

作者：Julian Lemmel,Felix Resch,Mónika Farsang,Ramin Hasani,Daniela Rus,Radu Grosu

点云|SLAM|雷达|激光|深度RGBD相关(3篇)

【1】MoDEx: Mixture of Depth-specific Experts for Multivariate Long-term Time Series Forecasting
标题：MoDEx：多元长期时间序列预测的特定深度专家混合
链接：https://arxiv.org/abs/2602.00624

作者：Hyekyung Yoon,Minhyuk Lee,Imseung Park,Myungjoo Kang

【2】Depth, Not Data: An Analysis of Hessian Spectral Bifurcation
标题：深度而非数据：Hessian谱分叉分析
链接：https://arxiv.org/abs/2602.00545

作者：Shenyang Deng,Boyao Liao,Zhuoli Ouyang,Tianyu Pang,Yaoqing Yang

【3】Real-Time 2D LiDAR Object Detection Using Three-Frame RGB Scan Encoding
标题：使用三帧Ruby扫描编码的实时2D LiDART对象检测
链接：https://arxiv.org/abs/2602.02167

作者：Soheil Behnam Roudsari,Alexandre S. Brandão,Felipe N. Martins
备注：6 pages, 6 figures, submitted to IEEE SAS 2026

联邦学习|隐私保护|加密(7篇)

【1】Age-Aware Edge-Blind Federated Learning via Over-the-Air Aggregation
标题：通过空中聚合的感知边缘盲联邦学习
链接：https://arxiv.org/abs/2602.02469

作者：Ahmed M. Elshazly,Ahmed Arafa
备注：To appear in IEEE ICC 2026

【2】Conflict-Aware Client Selection for Multi-Server Federated Learning
标题：多服务器联合学习的预算感知客户端选择
链接：https://arxiv.org/abs/2602.02458

作者：Mingwei Hong,Zheng Lin,Zehang Lin,Lin Li,Miao Yang,Xia Du,Zihan Fang,Zhaolu Kang,Dianxin Luan,Shunzhi Zhu
备注：6 pages, 4 figures

【3】Rethinking LoRA for Data Heterogeneous Federated Learning: Subspace and State Alignment
标题：数据异构联邦学习的LoRA再思考：子空间和状态对齐
链接：https://arxiv.org/abs/2602.01746

作者：Hongyi Peng,Han Yu,Xiaoxiao Li,Qiang Yang

【4】Toward Enhancing Representation Learning in Federated Multi-Task Settings
标题：增强联邦多任务环境中的表示学习
链接：https://arxiv.org/abs/2602.01626

作者：Mehdi Setayesh,Mahdi Beitollahi,Yasser H. Khalil,Hongliang Li
备注：This paper has been accepted at ICLR 2026

【5】Federated Learning at the Forefront of Fairness: A Multifaceted Perspective
标题：联邦学习在公平的前沿：一个多方面的视角
链接：https://arxiv.org/abs/2602.00718

作者：Noorain Mukhtiar,Adnan Mahmood,Yipeng Zhou,Jian Yang,Jing Teng,Quan Z. Sheng
备注：7 pages (main content), 2 pages (references), Accepted and Published Proceedings of the 34th International Joint Conference on Artificial Intelligence (IJCAI). 2025

【6】Forecasting Energy Availability in Local Energy Communities via LSTM Federated Learning
标题：通过LSTM联邦学习预测当地能源社区的能源可用性
链接：https://arxiv.org/abs/2602.00694

作者：Fabio Turazza,Marcello Pietri,Natalia Selini Hadjidimitriou,Marco Mamei
备注：Published as a book chapter in the MEDES 2024 proceedings (Springer LNCS)

【7】Federated Learning With Individualized Privacy Through Client Sampling
标题：通过客户端抽样实现个性化隐私的联邦学习
链接：https://arxiv.org/abs/2501.17634

作者：Lucas Lange,Ole Borchardt,Erhard Rahm
备注：Accepted at 10th International Conference on Machine Learning Technologies (ICMLT 2025)

推理|分析|理解|解释(33篇)

【1】MentisOculi: Revealing the Limits of Reasoning with Mental Imagery
标题：MentisOculi：用心理想象揭示推理的局限性
链接：https://arxiv.org/abs/2602.02465

作者：Jana Zeller,Thaddäus Wiedemer,Fanfei Li,Thomas Klein,Prasanna Mayilvahanan,Matthias Bethge,Felix Wichmann,Ryan Cotterell,Wieland Brendel
备注：9 pages, 8 figures

【2】Didactic to Constructive: Turning Expert Solutions into Learnable Reasoning
标题：说教到建设性：将专家解决方案转化为可学习推理
链接：https://arxiv.org/abs/2602.02405

作者：Ethan Mendes,Jungsoo Park,Alan Ritter

【3】C-kNN-LSH: A Nearest-Neighbor Algorithm for Sequential Counterfactual Inference
标题：C-kNN-LSH：序列反事实推理的最近邻算法
链接：https://arxiv.org/abs/2602.02371

作者：Jing Wang,Jie Shen,Qiaomin Xie,Jeremy C Weiss

【4】Fat-Cat: Document-Driven Metacognitive Multi-Agent System for Complex Reasoning
标题：Fat-Cat：文档驱动的复杂推理元认知多智能体系统
链接：https://arxiv.org/abs/2602.02206

作者：Tong Yang,Yemin Wang,Chaoning Zhang,Aming Wu

【5】Think Dense, Not Long: Dynamic Decoupled Conditional Advantage for Efficient Reasoning
标题：思考密集，不长：高效推理的动态脱钩条件优势
链接：https://arxiv.org/abs/2602.02099

作者：Keqin Peng,Yuanxin Ouyang,Xuebo Liu,Zhiliang Tian,Ruijian Han,Yancheng Yuan,Liang Ding

【6】Small Generalizable Prompt Predictive Models Can Steer Efficient RL Post-Training of Large Reasoning Models
标题：小型可推广即时预测模型可以引导大型推理模型的高效RL后训练
链接：https://arxiv.org/abs/2602.01970

作者：Yun Qu,Qi Wang,Yixiu Mao,Heming Zou,Yuhang Jiang,Weijie Liu,Clive Bai,Kai Yang,Yangkun Chen,Saiyong Yang,Xiangyang Ji

【7】Geometric Analysis of Token Selection in Multi-Head Attention
标题：多头注意力中代币选择的几何分析
链接：https://arxiv.org/abs/2602.01893

作者：Timur Mudarisov,Mikhal Burtsev,Tatiana Petrova,Radu State

【8】Entropy-Guided Data-Efficient Training for Multimodal Reasoning Reward Models
标题：多模式推理奖励模型的信息量引导的数据高效训练
链接：https://arxiv.org/abs/2602.01884

作者：Shidong Yang,Tongwen Huang,Hao Wen,Yong Wang,Li Chen,Xiangxiang Chu

【9】DOGMA: Weaving Structural Information into Data-centric Single-cell Transcriptomics Analysis
标题：DOGMA：将结构信息编织到以数据为中心的单细胞转录组学分析中
链接：https://arxiv.org/abs/2602.01839

作者：Ru Zhang,Xunkai Li,Yaxin Deng,Sicheng Liu,Daohan Su,Qiangqiang Dai,Hongchao Qin,Rong-Hua Li,Guoren Wang,Jia Li
备注：12 pages, 4 figures

【10】Beyond Precision: Training-Inference Mismatch is an Optimization Problem and Simple LR Scheduling Fixes It
标题：超越精确：训练-推理不匹配是一个优化问题，简单的LR调度可以解决它
链接：https://arxiv.org/abs/2602.01826

作者：Yaxiang Zhang,Yingru Li,Jiacai Liu,Jiawei Xu,Ziniu Li,Qian Liu,Haoyuan Li

【11】RedVisor: Reasoning-Aware Prompt Injection Defense via Zero-Copy KV Cache Reuse
标题：RedVisor：通过零复制KV缓存重用进行推理感知提示注入防御
链接：https://arxiv.org/abs/2602.01795

作者：Mingrui Liu,Sixiao Zhang,Cheng Long,Kwok-Yan Lam
备注：under review

【12】PRISM: Parametrically Refactoring Inference for Speculative Sampling Draft Models
标题：PRism：投机抽样草案模型的参数化重构推理
链接：https://arxiv.org/abs/2602.01762

作者：Xuliang Wang,Yuetao Chen,Maochan Zhen,Fang Liu,Xinzhou Zheng,Xingwu Liu,Hong Xu,Ming Li

【13】Restoring Exploration after Post-Training: Latent Exploration Decoding for Large Reasoning Models
标题：训练后恢复探索：大型推理模型的潜在探索解码
链接：https://arxiv.org/abs/2602.01698

作者：Wenhui Tan,Fiorenzo Parascandolo,Enver Sangineto,Jianzhong Ju,Zhenbo Luo,Qian Cao,Rita Cucchiara,Ruihua Song,Jian Luan

【14】Beyond Dense States: Elevating Sparse Transcoders to Active Operators for Latent Reasoning
标题：超越稠密状态：将稀疏代码转换器提升为主动操作符以进行潜在推理
链接：https://arxiv.org/abs/2602.01695

作者：Yadong Wang,Haodong Chen,Yu Tian,Chuanxing Geng,Dong Liang,Xiang Chen

【15】Learning to Guide Local Search for MPE Inference in Probabilistic Graphical Models
标题：学习在概率图形模型中指导本地搜索MBE推理
链接：https://arxiv.org/abs/2602.01475

作者：Brij Malhotra,Shivvrat Arya,Tahrima Rahman,Vibhav Giridhar Gogate

【16】Theoretical Analysis of Measure Consistency Regularization for Partially Observed Data
标题：部分观测数据的度量一致性正规化理论分析
链接：https://arxiv.org/abs/2602.01437

作者：Yinsong Wang,Shahin Shahrampour

【17】MarkovScale: Towards Optimal Sequential Scaling at Inference Time
标题：MarkovScale：在推理时实现最佳顺序缩放
链接：https://arxiv.org/abs/2602.01120

作者：Youkang Wang,Jian Wang,Rubing Chen,Tianyi Zeng,Xiao-Yong Wei,Qing Li
备注：12 pages

【18】How Does Unfaithful Reasoning Emerge from Autoregressive Training? A Study of Synthetic Experiments
标题：自回归训练如何产生不忠实的推理？综合实验研究
链接：https://arxiv.org/abs/2602.01017

作者：Fuxin Wang,Amr Alazali,Yiqiao Zhong
备注：25 pages, 23 figures

【19】PyGALAX: An Open-Source Python Toolkit for Advanced Explainable Geospatial Machine Learning
标题：PyGALAX：用于高级可解释地理空间机器学习的开源Python工具包
链接：https://arxiv.org/abs/2602.00907

作者：Pingping Wang,Yihong Yuan,Lingcheng Li,Yongmei Lu

【20】Stable Time Series Prediction of Enterprise Carbon Emissions Based on Causal Inference
标题：基于因果推理的企业碳排放稳定时间序列预测
链接：https://arxiv.org/abs/2602.00775

作者：Zitao Hong,Zhen Peng,Xueping Liu

【21】A novel VAE-DML fusion framework for casual analysis of greenwashing in the mining industry
标题：用于采矿业洗绿随意分析的新型VAE-TLR融合框架
链接：https://arxiv.org/abs/2602.00774

作者：Yuxin Lu,Zhen Peng,Xiqiang Xia,Jie Wang

【22】From Pixels to Facts (Pix2Fact): Benchmarking Multi-Hop Reasoning for Fine-Grained Visual Fact Checking
标题：从像素到事实（Pix2Fact）：用于细粒度视觉事实检查的多跳推理基准
链接：https://arxiv.org/abs/2602.00593

作者：Yifan Jiang,Cong Zhang,Bofei Zhang,Yifan Yang,Bingzhang Wang,Yew-Soon Ong

【23】Do Latent-CoT Models Think Step-by-Step? A Mechanistic Study on Sequential Reasoning Tasks
标题：潜伏CoT模型是否循序渐进地思考？顺序推理任务的机制研究
链接：https://arxiv.org/abs/2602.00449

作者：Jia Liang,Liangming Pan
备注：20 pages, 14 figures

【24】DISK: Dynamic Inference SKipping for World Models
标题：DISK：世界模型的动态推理跳过
链接：https://arxiv.org/abs/2602.00440

作者：Anugunj Naman,Gaibo Zhang,Ayushman Singh,Yaguang Zhang

【25】Leveraging Textual-Cues for Enhancing Multimodal Sentiment Analysis by Object Recognition
标题：利用文本线索通过对象识别增强多模式情感分析
链接：https://arxiv.org/abs/2602.00360

作者：Sumana Biswas,Karen Young,Josephine Griffith

【26】Prototype-based Explainable Neural Networks with Channel-specific Reasoning for Geospatial Learning Tasks
标题：用于地理空间学习任务的基于原型的可解释神经网络，具有特定于学科的推理
链接：https://arxiv.org/abs/2602.00331

作者：Anushka Narayanan,Karianne J. Bergen
备注：submitted to Environmental Data Science (preprint)

【27】Analyzing Shapley Additive Explanations to Understand Anomaly Detection Algorithm Behaviors and Their Complementarity
标题：分析Shapley加法解释以了解异常检测算法行为及其互补性
链接：https://arxiv.org/abs/2602.00208

作者：Jordan Levy,Paul Saves,Moncef Garouani,Nicolas Verstaevel,Benoit Gaudou
备注：Accepted at Intelligent Data Analysis (IDA), 2026

【28】Reducing Memorisation in Generative Models via Riemannian Bayesian Inference
标题：通过Riemann Bayesian推理减少生成模型中的小型化
链接：https://arxiv.org/abs/2602.00199

作者：Johanna Marie Gegenfurtner,Albert Kjøller Jacobsen,Naima Elosegui Borras,Alejandro Valverde Mahou,Georgios Arvanitidis

【29】ECCO: Evidence-Driven Causal Reasoning for Compiler Optimization
标题：ECCO：证据驱动的因果推理以实现更好的优化
链接：https://arxiv.org/abs/2602.00087

作者：Haolin Pan,Lianghong Huang,Jinyuan Dong,Mingjie Xing,Yanjun Wu

【30】Scalable and Secure AI Inference in Healthcare: A Comparative Benchmarking of FastAPI and Triton Inference Server on Kubernetes
标题：医疗保健领域的可扩展和安全人工智能推理：Kubernetes上FastAPI和Triton推理服务器的比较基准测试
链接：https://arxiv.org/abs/2602.00053

作者：Ratul Ali
备注：2 pages, 2 figures, 1 table

【31】Importance Weighted Variational Inference without the Reparameterization Trick
标题：无需重新参数化技巧的重要性加权变分推理
链接：https://arxiv.org/abs/2602.01412

作者：Kamélia Daudel,Minh-Ngoc Tran,Cheng Zhang

【32】Hessian Spectral Analysis at Foundation Model Scale
标题：基础模型比例的黑森谱分析
链接：https://arxiv.org/abs/2602.00816

作者：Diego Granziol,Khurshid Juarev

【33】Bitcoin Price Prediction using Machine Learning and Combinatorial Fusion Analysis
标题：使用机器学习和组合融合分析预测比特币价格
链接：https://arxiv.org/abs/2602.00037

作者：Yuanhong Wu,Wei Ye,Jingyan Xu,D. Frank Hsu
备注：8 pages, 5 figures, 3 tables; Accepted to 2025 IEEE Conference on Artificial Intelligence (IEEE CAI)

检测相关(9篇)

【1】Prediction-Powered Risk Monitoring of Deployed Models for Detecting Harmful Distribution Shifts
标题：对已部署模型进行预测驱动的风险监控，以检测有害的分布变化
链接：https://arxiv.org/abs/2602.02229

作者：Guangyi Zhang,Yunlong Cai,Guanding Yu,Osvaldo Simeone

【2】AICD Bench: A Challenging Benchmark for AI-Generated Code Detection
标题：AICD Bench：人工智能生成代码检测的领先基准
链接：https://arxiv.org/abs/2602.02079

作者：Daniil Orel,Dilshod Azizov,Indraneil Paul,Yuxia Wang,Iryna Gurevych,Preslav Nakov

【3】Your AI-Generated Image Detector Can Secretly Achieve SOTA Accuracy, If Calibrated
标题：如果经过校准，您的人工智能生成图像检测器可以秘密实现SOTA准确性
链接：https://arxiv.org/abs/2602.01973

作者：Muli Yang,Gabriel James Goenawan,Henan Wang,Huaiyuan Qin,Chenghao Xu,Yanhua Yang,Fen Fang,Ying Sun,Joo-Hwee Lim,Hongyuan Zhu
备注：AAAI 2026. Code: https://github.com/muliyangm/AIGI-Det-Calib

【4】RAPT: Model-Predictive Out-of-Distribution Detection and Failure Diagnosis for Sim-to-Real Humanoid Robots
标题：RAPT：Sim-to-Real人形机器人的模型预测分布失调检测和故障诊断
链接：https://arxiv.org/abs/2602.01515

作者：Humphrey Munn,Brendan Tidd,Peter Bohm,Marcus Gallagher,David Howard

【5】PaAno: Patch-Based Representation Learning for Time-Series Anomaly Detection
标题：PaAno：用于时间序列异常检测的基于补丁的表示学习
链接：https://arxiv.org/abs/2602.01359

作者：Jinju Park,Seokho Kang
备注：Accepted by the 14th International Conference on Learning Representations (ICLR 2026)

【6】Do Schwartz Higher-Order Values Help Sentence-Level Human Value Detection? When Hard Gating Hurts
标题：施瓦茨更高级价值观有助于句子级的人类价值检测吗？当硬门伤人时
链接：https://arxiv.org/abs/2602.00913

作者：Víctor Yeste,Paolo Rosso
备注：Code: https://github.com/VictorMYeste/human-value-detection, 42 pages, 4 figures

【7】Detecting AI-Generated Content in Academic Peer Reviews
标题：在学术同行评审中检测人工智能生成的内容
链接：https://arxiv.org/abs/2602.00319

作者：Siyuan Shen,Kai Wang

【8】GEPC: Group-Equivariant Posterior Consistency for Out-of-Distribution Detection in Diffusion Models
标题：GEPC：扩散模型中分布外检测的群等变后验一致性
链接：https://arxiv.org/abs/2602.00191

作者：Yadang Alexis Rouzoumka,Jean Pinsolle,Eugénie Terreaux,Christèle Morisseau,Jean-Philippe Ovarlez,Chengfang Ren
备注：preprint

【9】Comparison of Multiple Classifiers for Android Malware Detection with Emphasis on Feature Insights Using CICMalDroid 2020 Dataset
标题：使用CICMalDroid 2020数据集比较Android恶意软件检测的多个分类器，重点关注功能洞察
链接：https://arxiv.org/abs/2602.00058

作者：Md Min-Ha-Zul Abedin,Tazqia Mehrub

分类|识别(12篇)

【1】Generalized Optimal Classification Trees: A Mixed-Integer Programming Approach
标题：广义最优分类树：混合子菜单规划方法
链接：https://arxiv.org/abs/2602.02173

作者：Jiancheng Tu,Wenqi Fan,Zhibin Wu

【2】Rethinking Genomic Modeling Through Optical Character Recognition
标题：通过光学字符识别重新思考基因组建模
链接：https://arxiv.org/abs/2602.02014

作者：Hongxin Xiang,Pengsen Ma,Yunkang Cao,Di Yu,Haowen Chen,Xinyu Yang,Xiangxiang Zeng

【3】Rotation-free Online Handwritten Character Recognition Using Linear Recurrent Units
标题：基于线性递归单元的无旋转在线手写体字符识别
链接：https://arxiv.org/abs/2602.01533

作者：Zhe Ling,Sicheng Yu,Danyu Yang

【4】Early Classification of Time Series in Non-Stationary Cost Regimes
标题：非平稳成本制度中时间序列的早期分类
链接：https://arxiv.org/abs/2602.00918

作者：Aurélien Renault,Alexis Bondu,Antoine Cornuéjols,Vincent Lemaire

【5】Mobile Exergames: Activity Recognition Based on Smartphone Sensors
标题：移动收件箱游戏：基于智能手机传感器的活动识别
链接：https://arxiv.org/abs/2602.00809

作者：David Craveiro,Hugo Silva

【6】Three-Way Emotion Classification of EEG-based Signals using Machine Learning
标题：使用机器学习对基于脑电的信号进行三向情绪分类
链接：https://arxiv.org/abs/2602.00670

作者：Ashna Purwar,Gaurav Simkar,Madhumita,Sachin Kadam
备注：6 pages, 8 figures, and 3 tables. Submitted to a conference, under review

【7】When Classes Evolve: A Benchmark and Framework for Stage-Aware Class-Incremental Learning
标题：当类演变：阶段感知类增量学习的基准和框架
链接：https://arxiv.org/abs/2602.00573

作者：Zheng Zhang,Tao Hu,Xueheng Li,Yang Wang,Rui Li,Jie Zhang,Chengjun Xie

【8】Contrastive Domain Generalization for Cross-Instrument Molecular Identification in Mass Spectrometry
标题：跨分子仪器鉴定的对比域概括
链接：https://arxiv.org/abs/2602.00547

作者：Seunghyun Yoo,Sanghong Kim,Namkyung Yoon,Hwangnam Kim
备注：8 pages, 2 figures

【9】Partition of Unity Neural Networks for Interpretable Classification with Explicit Class Regions
标题：具有显式类别区域的可解释分类的统一神经网络划分
链接：https://arxiv.org/abs/2602.00511

作者：Akram Aldroubi

【10】Quantum Phase Recognition via Quantum Attention Mechanism
标题：通过量子注意力机制的量子阶段识别
链接：https://arxiv.org/abs/2602.00473

作者：Jin-Long Chen,Xin Li,Zhang-Qi Yin
备注：10 pages, 7 figures

【11】Neuron Block Dynamics for XOR Classification with Zero-Margin
标题：零余量异或分类的神经元块动力学
链接：https://arxiv.org/abs/2602.00172

作者：Guillaume Braun,Masaaki Imaizumi
备注：47 pages, 9 figures

【12】Comparison of Image Processing Models in Quark Gluon Jet Classification
标题：量子胶子喷流分类中图像处理模型的比较
链接：https://arxiv.org/abs/2602.00141

作者：Daeun Kim,Jiwon Lee,Wonjun Jeong,Hyeongwoo Noh,Giyeong Kim,Jaeyoon Cho,Geonhee Kwak,Seunghwan Yang,MinJung Kweon
备注：17 pages, 10 Figures

表征(9篇)

【1】Implicit neural representation of textures
标题：纹理的隐式神经表示
链接：https://arxiv.org/abs/2602.02354

作者：Albert Kwok,Zheyuan Hu,Dounia Hammou
备注：Albert Kwok and Zheyuan Hu contributed equally to this work

【2】VQ-Style: Disentangling Style and Content in Motion with Residual Quantized Representations
标题：VQ风格：用残余量化表示解开运动中的风格和内容
链接：https://arxiv.org/abs/2602.02334

作者：Fatemeh Zargarbashi,Dhruv Agrawal,Jakob Buhmann,Martin Guay,Stelian Coros,Robert W. Sumner

【3】Learning Sparse Visual Representations via Spatial-Semantic Factorization
标题：通过空间-语义分解学习稀疏视觉表示
链接：https://arxiv.org/abs/2602.01905

作者：Theodore Zhengde Zhao,Sid Kiblawi,Jianwei Yang,Naoto Usuyama,Reuben Tan,Noel C Codella,Tristan Naumann,Hoifung Poon,Mu Wei

【4】ToPT: Task-Oriented Prompt Tuning for Urban Region Representation Learning
标题：ToPT：城市区域表示学习的面向任务的即时调整
链接：https://arxiv.org/abs/2602.01610

作者：Zitao Guo,Changyang Jiang,Tianhong Zhao,Jinzhou Cao,Genan Dai,Bowen Zhang
备注：The paper has been accepted by ICASSP 2026

【5】The Inlet Rank Collapse in Implicit Neural Representations: Diagnosis and Unified Remedy
标题：内隐神经表示中的入口等级崩溃：诊断和统一补救措施
链接：https://arxiv.org/abs/2602.01526

作者：Jianqiao Zheng,Hemanth Saratchandran,Simon Lucey

【6】Rectified LpJEPA: Joint-Embedding Predictive Architectures with Sparse and Maximum-Entropy Representations
标题：纠正的LpJEPA：具有稀疏和最大熵表示的联合嵌入预测架构
链接：https://arxiv.org/abs/2602.01456

作者：Yilun Kuang,Yash Dagade,Tim G. J. Rudner,Randall Balestriero,Yann LeCun

【7】Convergent World Representations and Divergent Tasks
标题：趋同的世界代表和分歧的任务
链接：https://arxiv.org/abs/2602.00533

作者：Core Francisco Park

【8】Localized, High-resolution Geographic Representations with Slepian Functions
标题：具有Slepian功能的本地化、高分辨率地理表示
链接：https://arxiv.org/abs/2602.00392

作者：Arjun Rao,Ruth Crasto,Tessa Ooms,David Rolnick,Konstantin Klemmer,Marc Rußwurm
备注：23 pages, 12 figures, 6 tables

【9】On the Relationship Between Representation Geometry and Generalization in Deep Neural Networks
标题：深度神经网络中表示几何与推广之间的关系
链接：https://arxiv.org/abs/2602.00130

作者：Sumit Yadav
备注：ICML

3D|3D重建等相关(3篇)

【1】CLAMP: Contrastive Learning for 3D Multi-View Action-Conditioned Robotic Manipulation Pretraining
标题：CLAMP：3D多视图条件机器人操纵预训练的对比学习
链接：https://arxiv.org/abs/2602.00937

作者：I-Chun Arthur Liu,Krzysztof Choromanski,Sandy Huang,Connor Schenck

【2】3DGS$^2$-TR: Scalable Second-Order Trust-Region Method for 3D Gaussian Splatting
标题：3DGS$^2$-TR：三维高斯溅射的可扩展二阶信赖域方法
链接：https://arxiv.org/abs/2602.00395

作者：Roger Hsiao,Yuchen Fang,Xiangru Huang,Ruilong Li,Hesam Rabeti,Zan Gojcic,Javad Lavaei,James Demmel,Sophia Shao

【3】Opportunistic Promptable Segmentation: Leveraging Routine Radiological Annotations to Guide 3D CT Lesion Segmentation
标题：启发式可分割分割：利用常规放射学注释来指导3D CT病变分割
链接：https://arxiv.org/abs/2602.00309

作者：Samuel Church,Joshua D. Warner,Danyal Maqbool,Xin Tie,Junjie Hu,Meghan G. Lubner,Tyler J. Bradshaw

编码器(4篇)

【1】You Need an Encoder for Native Position-Independent Caching
标题：您需要一个编码器来进行本地位置独立缓存
链接：https://arxiv.org/abs/2602.01519

作者：Shiju Zhao,Junhao Hu,Jiaqi Zheng,Guihai Chen
备注：12 pages, 10 figures. Welcome back, Encoder

【2】Where to Attend: A Principled Vision-Centric Position Encoding with Parabolas
标题：参加哪里：有原则的以视觉为中心的位置编码，使用Parabolas
链接：https://arxiv.org/abs/2602.01418

作者：Christoffer Koo Øhrstrøm,Rafael I. Cabral Muchacho,Yifei Dong,Filippos Moumtzidellis,Ronja Güldenring,Florian T. Pokorny,Lazaros Nalpantidis

【3】The Stacked Autoencoder Evolution Hypothesis
标题：堆叠自动编码器进化假说
链接：https://arxiv.org/abs/2602.01026

作者：Hiroyuki Iizuka

【4】Artificial Intelligence and Symmetries: Learning, Encoding, and Discovering Structure in Physical Data
标题：人工智能和对称性：学习、编码和发现物理数据中的结构
链接：https://arxiv.org/abs/2602.02351

作者：Veronica Sanz
备注：25 pages, 9 figures. This manuscript is an invited review at the International Journal of Modern Physics A

优化|敛散性(29篇)

【1】Maximizing Reliability with Bayesian Optimization
标题：用贝叶斯优化最大化可靠性
链接：https://arxiv.org/abs/2602.02432

作者：Jack M. Buckingham,Ivo Couckuyt,Juergen Branke
备注：25 pages, 9 figures

【2】SLIME: Stabilized Likelihood Implicit Margin Enforcement for Preference Optimization
标题：SlIME：稳定的潜在隐性保证金强制执行以实现偏好优化
链接：https://arxiv.org/abs/2602.02383

作者：Maksim Afanasyev,Illarion Iov

【3】An Optimization Method for Autoregressive Time Series Forecasting
标题：自回归时间序列预测的优化方法
链接：https://arxiv.org/abs/2602.02288

作者：Zheng Li,Jerry Cheng,Huanying Gu
备注：10 pages, 2 figures, 2 tables

【4】Variational Entropic Optimal Transport
标题：变分熵最优运输
链接：https://arxiv.org/abs/2602.02241

作者：Roman Dyachenko,Nikita Gushchin,Kirill Sokolov,Petr Mokrov,Evgeny Burnaev,Alexander Korotin

【5】Autocorrelated Optimize-via-Estimate: Predict-then-Optimize versus Finite-sample Optimal
标题：通过估计自相关优化：先预测然后优化与伪样本最佳
链接：https://arxiv.org/abs/2602.01877

作者：Zichun Wang,Gar Goei Loke,Ruiting Zuo

【6】FUPareto: Bridging the Forgetting-Utility Gap in Federated Unlearning via Pareto Augmented Optimization
标题：FUPareto：通过Pareto增强优化弥合联邦取消学习中的遗忘与效用差距
链接：https://arxiv.org/abs/2602.01852

作者：Zeyan Wang,Zhengmao Liu,Yongxin Cai,Chi Li,Xiaoying Tang,Jingchao Chen,Zibin Pan,Jing Qiu

【7】Cost-Aware Bayesian Optimization for Prototyping Interactive Devices
标题：交互式设备原型设计的成本意识Bayesian优化
链接：https://arxiv.org/abs/2602.01774

作者：Thomas Langerak,Renate Zhang,Ziyuan Wang,Per Ola Kristensson,Antti Oulasvirta

【8】Finite and Corruption-Robust Regret Bounds in Online Inverse Linear Optimization under M-Convex Action Sets
标题：M-凸动作集下在线逆线性优化中的有限和腐蚀鲁棒遗憾界
链接：https://arxiv.org/abs/2602.01682

作者：Taihei Oki,Shinsaku Sakaue

【9】Optimal Sample Complexity for Single Time-Scale Actor-Critic with Momentum
标题：具有动量的单时间尺度演员评论家的最佳样本复杂性
链接：https://arxiv.org/abs/2602.01505

作者：Navdeep Kumar,Tehila Dahan,Lior Cohen,Ananyabrata Barua,Giorgia Ramponi,Kfir Yehuda Levy,Shie Mannor

【10】Finding Differentially Private Second Order Stationary Points in Stochastic Minimax Optimization
标题：随机极小优化中寻找不同的私有二阶稳定点
链接：https://arxiv.org/abs/2602.01339

作者：Difei Xu,Youming Tao,Meng Ding,Chenglin Fan,Di Wang

【11】Rethinking the Flow-Based Gradual Domain Adaption: A Semi-Dual Optimal Transport Perspective
标题：重新思考基于流量的渐进领域适应：半二元最佳运输视角
链接：https://arxiv.org/abs/2602.01179

作者：Zhichao Chen,Zhan Zhuang,Yunfei Teng,Hao Wang,Fangyikang Wang,Zhengnan Li,Tianqiao Liu,Haoxuan Li,Zhouchen Lin

【12】PDE-Constrained Optimization for Neural Image Segmentation with Physics Priors
标题：具有物理先验的PED约束神经图像分割优化
链接：https://arxiv.org/abs/2602.01069

作者：Seema K. Poudel,Sunny K. Khadka

【13】Error Taxonomy-Guided Prompt Optimization
标题：错误分类引导的提示优化
链接：https://arxiv.org/abs/2602.00997

作者：Mayank Singh,Vikas Yadav,Eduardo Blanco

【14】Scalable Random Wavelet Features: Efficient Non-Stationary Kernel Approximation with Convergence Guarantees
标题：可扩展随机子波特征：具有收敛保证的高效非平稳核逼近
链接：https://arxiv.org/abs/2602.00987

作者：Sawan Kumar,Souvik Chakraborty
备注：Accepted at ICLR 2026

【15】Continuous-Utility Direct Preference Optimization
标题：连续效用直接偏好优化
链接：https://arxiv.org/abs/2602.00931

作者：Muhammad Ahmed Mohsin,Muhammad Umer,Ahsan Bilal,Zihao He,Muhammad Usman Rafique,Asad Aali,Muhammad Ali Jamshed,John M. Cioffi,Emily Fox
备注：Submitted to ICML 2026

【16】Hallucination is a Consequence of Space-Optimality: A Rate-Distortion Theorem for Membership Testing
标题：幻觉是空间最优性的后果：隶属度测试的速率失真定理
链接：https://arxiv.org/abs/2602.00906

作者：Anxin Guo,Jingwei Li

【17】Multi-Objective Multi-Fidelity Bayesian Optimization with Causal Priors
标题：具有因果先验的多目标多保真度贝叶斯优化
链接：https://arxiv.org/abs/2602.00788

作者：Md Abir Hossen,Mohammad Ali Javidian,Vignesh Narayanan,Jason M. O'Kane,Pooyan Jamshidi

【18】Pareto-Conditioned Diffusion Models for Offline Multi-Objective Optimization
标题：离线多目标优化的帕累托条件扩散模型
链接：https://arxiv.org/abs/2602.00737

作者：Jatan Shrestha,Santeri Heiskanen,Kari Hepola,Severi Rissanen,Pekka Jääskeläinen,Joni Pajarinen
备注：Accepted by ICLR 2026. Project page: https://sites.google.com/view/pcd-iclr26

【19】Combinatorial Bandit Bayesian Optimization for Tensor Outputs
标题：张量输出的组合Bandit Bayesian优化
链接：https://arxiv.org/abs/2602.00640

作者：Jingru Huang,Haijie Xu,Jie Guo,Manrui Jiang,Chen Zhang

【20】Direct Preference Optimization with Rating Information: Practical Algorithms and Provable Gains
标题：利用评级信息进行直接偏好优化：实用算法和可证明的收益
链接：https://arxiv.org/abs/2602.00603

作者：Luca Viano,Ruida Zhou,Yifan Sun,Mahdi Namazifar,Volkan Cevher,Shoham Sabach,Mohammad Ghavamzadeh

【21】Surrogate Ensemble in Expensive Multi-Objective Optimization via Deep Q-Learning
标题：通过深度Q学习替代昂贵的多目标优化
链接：https://arxiv.org/abs/2602.00540

作者：Yuxin Wu,Hongshu Guo,Ting Huang,Yue-Jiao Gong,Zeyuan Ma

【22】Quality-Diversity Optimization as Multi-Objective Optimization
标题：作为多目标优化的质量多样性优化
链接：https://arxiv.org/abs/2602.00478

作者：Xi Lin,Ping Guo,Yilu Liu,Qingfu Zhang,Jianyong Sun

【23】Diffusion LMs Can Approximate Optimal Infilling Lengths Implicitly
标题：扩散LM可以逼近最佳填充长度
链接：https://arxiv.org/abs/2602.00476

作者：Hengchang Liu,Zhao Yang,Bing Su

【24】Dimensional Peeking for Low-Variance Gradients in Zeroth-Order Discrete Optimization via Simulation
标题：零阶离散优化中低方差要素的维度透视
链接：https://arxiv.org/abs/2602.00075

作者：Philipp Andelfinger,Wentong Cai
备注：Accepted at ACM SIGSIM PADS 2026

【25】TextBFGS: Quasi-Newton Optimization for Discrete Executable Text via Gradient-Operator Retrieval
标题：文本BFSG：通过发件人-操作员检索对离散可执行文本进行拟牛顿优化
链接：https://arxiv.org/abs/2602.00059

作者：Zizheng Zhang,Yuyang Liao,Chen Chen,Jian He,Dun Wu,Qianjin Yu,Yanqin Gao,Jin Yang,Kailai Zhang,Eng Siong Chng,Xionghu Zhong

【26】Minimax optimal differentially private synthetic data for smooth queries
标题：Minimax最佳差异私密合成数据，以实现平滑查询
链接：https://arxiv.org/abs/2602.01607

作者：Rundong Ding,Yiyun He,Yizhe Zhu
备注：27 pages

【27】Robust Sublinear Convergence Rates for Iterative Bregman Projections
标题：迭代Bregman投影的鲁棒次线性收敛率
链接：https://arxiv.org/abs/2602.01372

作者：Gabriel Peyré

【28】Optimal Decision-Making Based on Prediction Sets
标题：基于预测集的最优决策
链接：https://arxiv.org/abs/2602.00989

作者：Tao Wang,Edgar Dobriban

【29】On the Convergence of Jacobian-Free Backpropagation for Optimal Control Problems with Implicit Hamiltonians
标题：带隐Hamilton最优控制问题的无Jacobian反向传播的收敛性
链接：https://arxiv.org/abs/2602.00921

作者：Eric Gelphman,Deepanshu Verma,Nicole Tianjiao Yang,Stanley Osher,Samy Wu Fung
备注：19 Pages, 6 figures, 1 table. Submitted to ICML and is pending review

预测|估计(32篇)

【1】Back to the Future: Look-ahead Augmentation and Parallel Self-Refinement for Time Series Forecasting
标题：回到未来：时间序列预测的前瞻增强和并行自我细化
链接：https://arxiv.org/abs/2602.02146

作者：Sunho Kim,Susik Yoon
备注：4 pages, Short paper accepted at The Web Conference (WWW) 2026

【2】Embedding Learning on Multiplex Networks for Link Prediction
标题：在多重网络上嵌入学习以进行链路预测
链接：https://arxiv.org/abs/2602.01922

作者：Orell Trautmann,Olaf Wolkenhauer,Clémence Réda

【3】Stein-Rule Shrinkage for Stochastic Gradient Estimation in High Dimensions
标题：多维随机梯度估计的斯坦规则收缩
链接：https://arxiv.org/abs/2602.01777

作者：M. Arashi,M. Amintoosi

【4】Position: Beyond Model-Centric Prediction -- Agentic Time Series Forecasting
标题：位置：超越以模型为中心的预测--统计时间序列预测
链接：https://arxiv.org/abs/2602.01776

作者：Mingyue Cheng,Xiaoyu Tao,Qi Liu,Ze Guo,Enhong Chen

【5】Position: The Inevitable End of One-Architecture-Fits-All-Domains in Time Series Forecasting
标题：位置：时间序列预测中一个架构适合所有领域的必然结束
链接：https://arxiv.org/abs/2602.01736

作者：Qinwei Ma,Jingzhe Shi,Jiahao Qiu,Zaiwen Yang
备注：14 pages, 3 figures, 2 tables

【6】SafePred: A Predictive Guardrail for Computer-Using Agents via World Models
标题：SafePred：通过世界模型为使用计算机的代理提供预测保障
链接：https://arxiv.org/abs/2602.01725

作者：Yurun Chen,Zeyi Liao,Ping Yin,Taotao Xie,Keting Yin,Shengyu Zhang

【7】AgroFlux: A Spatial-Temporal Benchmark for Carbon and Nitrogen Flux Prediction in Agricultural Ecosystems
标题：AgroFlux：农业生态系统碳氮通量预测的时空基准
链接：https://arxiv.org/abs/2602.01614

作者：Qi Cheng,Licheng Liu,Yao Zhang,Mu Hong,Yiqun Xie,Xiaowei Jia

【8】Spectral Text Fusion: A Frequency-Aware Approach to Multimodal Time-Series Forecasting
标题：谱文本融合：一种多峰时间序列预测的频率感知方法
链接：https://arxiv.org/abs/2602.01588

作者：Huu Hiep Nguyen,Minh Hoang Nguyen,Dung Nguyen,Hung Le

【9】A Lightweight Sparse Interaction Network for Time Series Forecasting
标题：用于时间序列预测的轻量级稀疏交互网络
链接：https://arxiv.org/abs/2602.01585

作者：Xu Zhang,Qitong Wang,Peng Wang,Wei Wang
备注：The paper is published in AAAI Conference on Artificial Intelligence, AAAI 2025. The code is available at the link https://github.com/Meteor-Stars/LSINet

【10】Predicting and improving test-time scaling laws via reward tail-guided search
标题：通过奖励尾引导搜索预测和改进测试时间缩放定律
链接：https://arxiv.org/abs/2602.01485

作者：Muheng Li,Jian Qian,Wenlong Mou
备注：33 pages, 5 figures

【11】An Odd Estimator for Shapley Values
标题：沙普利值的奇怪估计
链接：https://arxiv.org/abs/2602.01399

作者：Fabian Fumagalli,Landon Butler,Justin Singh Kang,Kannan Ramchandran,R. Teal Witter

【12】Deep Variational Contrastive Learning for Joint Risk Stratification and Time-to-Event Estimation
标题：用于联合风险分层和事件时间估计的深度变分对比学习
链接：https://arxiv.org/abs/2602.01367

作者：Pinar Erbil,Alberto Archetti,Eugenio Lomurno,Matteo Matteucci

【13】Multi-Horizon Electricity Price Forecasting with Deep Learning in the Australian National Electricity Market
标题：澳大利亚国家电力市场利用深度学习进行多水平电价预测
链接：https://arxiv.org/abs/2602.01157

作者：Mohammed Osman Gani,Zhipeng He,Chun Ouyang,Sara Khalifa
备注：63 Pages

【14】TRACE: Scalable Amortized Causal Discovery from Single Sequences via Autoregressive Density Estimation
标题：TRACE：通过自回归密度估计从单个序列中发现可扩展的摊销因果关系
链接：https://arxiv.org/abs/2602.01135

作者：Hugo Math,Rainer Lienhart
备注：8 pages, 6 figures,

【15】Predicting Anemia Among Under-Five Children in Nepal Using Machine Learning and Deep Learning
标题：使用机器学习和深度学习预测尼泊尔五岁以下儿童的贫血
链接：https://arxiv.org/abs/2602.01005

作者：Deepak Bastola,Pitambar Acharya,Dipak Dulal,Rabina Dhakal,Yang Li
备注：13 pages and submission to Public Health Nutrition is in progress

【16】Don't Forget Its Variance! The Minimum Path Variance Principle for Accurate and Stable Score-Based Density Ratio Estimation
标题：不要忘记它的差异！准确稳定的基于分数的密度比估计的最小路径方差原则
链接：https://arxiv.org/abs/2602.00834

作者：Wei Chen,Jiacheng Li,Shigui Li,Zhiqi Lin,Junmei Yang,John Paisley,Delu Zeng

【17】Deep Time-series Forecasting Needs Kernelized Moment Balancing
标题：深度时间序列预测需要核心化的时刻平衡
链接：https://arxiv.org/abs/2602.00717

作者：Licheng Pan,Hao Wang,Haocheng Yang,Yuqi Li,Qingsong Wen,Xiaoxi Li,Zhichao Chen,Haoxuan Li,Zhixuan Chu,Yuan Lu

【18】PHAT: Modeling Period Heterogeneity for Multivariate Time Series Forecasting
标题：PHAT：多元时间序列预测的周期异方差建模
链接：https://arxiv.org/abs/2602.00654

作者：Jiaming Ma,Guanjun Wang,Qihe Huang,Sheng Huang,Haofeng Ma,Zhengyang Zhou,Pengkun Wang,Binwu Wang,Yang Wang

【19】Bridging Time and Frequency: A Joint Modeling Framework for Irregular Multivariate Time Series Forecasting
标题：时间与频率的桥梁：不规则多元时间序列预测的联合建模框架
链接：https://arxiv.org/abs/2602.00582

作者：Xiangfei Qiu,Kangjia Yan,Xvyuan Liu,Xingjian Wu,Jilin Hu

【20】OpenDDI: A Comprehensive Benchmark for DDI Prediction
标题：OpenDID：DDD预测的综合基准
链接：https://arxiv.org/abs/2602.00539

作者：Xinmo Jin,Bowen Fan,Xunkai Li,Henan Sun,YuXin Zeng,Zekai Chen,Yuxuan Sun,Jia Li,Qiangqiang Dai,Hongchao Qin,Rong-Hua Li,Guoren Wang

【21】PAIR-Former: Budgeted Relational MIL for miRNA Target Prediction
标题：PAIR-Former：用于miRNA靶点预测的预定关系MIL
链接：https://arxiv.org/abs/2602.00465

作者：Jiaqi Yin,Baiming Chen,Jia Fei,Mingjun Yang
备注：Preprint. Under review. During the preprint stage, inquiries and feedback can be directed to Jiaqi Yin (yjqhit@gmail.com)

【22】From Observations to States: Latent Time Series Forecasting
标题：从观察到状态：潜在时间序列预测
链接：https://arxiv.org/abs/2602.00297

作者：Jie Yang,Yifan Hu,Yuante Li,Kexin Zhang,Kaize Ding,Philip S. Yu

【23】Green-NAS: A Global-Scale Multi-Objective Neural Architecture Search for Robust and Efficient Edge-Native Weather Forecasting
标题：Green-NAS：全球规模多目标神经架构搜索，用于强大且高效的边缘原生天气预报
链接：https://arxiv.org/abs/2602.00240

作者：Md Muhtasim Munif Fahim,Soyda Humyra Yesmin,Saiful Islam,Md. Palash Bin Faruque,Md. A. Salam,Md. Mahfuz Uddin,Samiul Islam,Tofayel Ahmed,Md. Binyamin,Md. Rezaul Karim

【24】Predicting Mortgage Default with Machine Learning: AutoML, Class Imbalance, and Leakage Control
标题：利用机器学习预测抵押贷款违约：AutoML、阶级失衡和泄漏控制
链接：https://arxiv.org/abs/2602.00120

作者：Xianghong Hu,Tianning Xu,Ying Chen,Shuai Wang
备注：12 pages, 4 figures. An extended and pedagogical version will appear as a book chapter

【25】Automated univariate time series forecasting with regression trees
标题：利用回归树自动化单变量时间序列预测
链接：https://arxiv.org/abs/2602.00077

作者：Francisco Martínez,María P. Frías
备注：23 pages, 17 figures

【26】Disentangled Interest Network for Out-of-Distribution CTR Prediction
标题：解纠缠兴趣网络用于发行外点击率预测
链接：https://arxiv.org/abs/2602.00002

作者：Yu Zheng,Chen Gao,Jianxin Chang,Yanan Niu,Yang Song,Depeng Jin,Meng Wang,Yong Li
备注：Accepted by ACM TOIS

【27】Reliable Real-Time Value at Risk Estimation via Quantile Regression Forest with Conformal Calibration
标题：通过分位数回归森林和保形校准进行可靠的实时风险价值估计
链接：https://arxiv.org/abs/2602.01912

作者：Du-Yi Wang,Guo Liang,Kun Zhang,Qianwen Zhu

【28】ST-BCP: Tightening Coverage Bound for Backward Conformal Prediction via Non-Conformity Score Transformation
标题：ST-BEP：通过不合格评分转换收紧后向保形预测的覆盖范围
链接：https://arxiv.org/abs/2602.01733

作者：Junxian Liu,Hao Zeng,Hongxin Wei

【29】Non-Uniform Noise-to-Signal Ratio in the REINFORCE Policy-Gradient Estimator
标题：REINFORCE策略梯度估计中的非均匀噪信比
链接：https://arxiv.org/abs/2602.01460

作者：Haoyu Han,Heng Yang

【30】Improving Minimax Estimation Rates for Contaminated Mixture of Multinomial Logistic Experts via Expert Heterogeneity
标题：通过专家异方差提高受污染的多项逻辑专家混合物的极小极大估计率
链接：https://arxiv.org/abs/2602.00939

作者：Fanqi Yan,Dung Le,Trang Pham,Huy Nguyen,Nhat Ho
备注：Fanqi Yan, Dung Le contributed equally to this work. 41 pages, 3 figures, 1 table

【31】Early warning prediction: Onsager-Machlup vs Schrödinger
标题：预警预测：昂萨格-马赫卢普vs薛定格
链接：https://arxiv.org/abs/2602.00143

作者：Xiaoai Xu,Yixuan Zhou,Xiang Zhou,Jingqiao Duan,Ting Gao
备注：20 pages

【32】Exploring the Interpretability of Forecasting Models for Energy Balancing Market
标题：探索能源平衡市场预测模型的可解释性
链接：https://arxiv.org/abs/2602.00049

作者：Oskar Våle,Shiliang Zhang,Sabita Maharjan,Gro Klæboe
备注：Accepted by Artificial Intelligence Science and Engineering. Copyright has been transferred to IEEE

其他神经网络|深度学习|模型|建模(90篇)

【1】RLAnything: Forge Environment, Policy, and Reward Model in Completely Dynamic RL System
标题：RL Anything：完全动态RL系统中的锻造环境、政策和奖励模型
链接：https://arxiv.org/abs/2602.02488

作者：Yinjie Wang,Tianbao Xie,Ke Shen,Mengdi Wang,Ling Yang
备注：Code: https://github.com/Gen-Verse/Open-AgentRL

【2】SPARKLING: Balancing Signal Preservation and Symmetry Breaking for Width-Progressive Learning
标题：SPARKLING：平衡信号保留和对称性破坏以实现广度渐进学习
链接：https://arxiv.org/abs/2602.02472

作者：Qifan Yu,Xinyu Ma,Zhijian Zhuo,Minrui Wang,Deyi Liu,Shiyi Zhan,Yiyuan Ma,Liang Xiang,Xingyan Bin,Di He

【3】PRISM: Performer RS-IMLE for Single-pass Multisensory Imitation Learning
标题：PRism：用于单次多感官模仿学习的表演者RS-IMLE
链接：https://arxiv.org/abs/2602.02396

作者：Amisha Bhaskar,Pratap Tokekar,Stefano Di Cairano,Alexander Schperberg
备注：10 pages main text and 4 figures, and 11 pages appendix and 10 figures, total 21 pages and 14 figures

【4】Context Learning for Multi-Agent Discussion
标题：用于多智能体讨论的上下文学习
链接：https://arxiv.org/abs/2602.02350

作者：Xingyuan Hua,Sheng Yue,Xinyi Li,Yizhe Zhao,Jinrui Zhang,Ju Ren

【5】Spark: Modular Spiking Neural Networks
标题：Spark：模块化尖峰神经网络
链接：https://arxiv.org/abs/2602.02306

作者：Mario Franco,Carlos Gershenson

【6】Decoupling Generalizability and Membership Privacy Risks in Neural Networks
标题：神经网络中的可概括性和成员隐私风险脱钩
链接：https://arxiv.org/abs/2602.02296

作者：Xingli Fang,Jung-Eun Kim

【7】Statistical Learning Theory in Lean 4: Empirical Processes from Scratch
标题：精益4中的统计学习理论：Scratch的经验过程
链接：https://arxiv.org/abs/2602.02285

作者：Yuanhe Zhang,Jason D. Lee,Fanghui Liu
备注：19 pages, 2 figures. Comments are welcome

【8】Choice-Model-Assisted Q-learning for Delayed-Feedback Revenue Management
标题：延迟反馈收入管理的选择模型辅助Q学习
链接：https://arxiv.org/abs/2602.02283

作者：Owen Shen,Patrick Jaillet

【9】Learning Markov Decision Processes under Fully Bandit Feedback
标题：完全Bandit反馈下的Markov决策过程学习
链接：https://arxiv.org/abs/2602.02260

作者：Zhengjia Zhuo,Anupam Gupta,Viswanath Nagarajan

【10】Segment to Focus: Guiding Latent Action Models in the Presence of Distractors
标题：细分到焦点：在干扰因素存在的情况下引导潜在行动模式
链接：https://arxiv.org/abs/2602.02259

作者：Hamza Adnan,Matthew T. Jackson,Alexey Zakharov

【11】Interpretability in Deep Time Series Models Demands Semantic Alignment
标题：深度时间序列模型的可解释性需要语义一致
链接：https://arxiv.org/abs/2602.02239

作者：Giovanni De Felice,Riccardo D'Elia,Alberto Termine,Pietro Barbiero,Giuseppe Marra,Silvia Santini

【12】SurvKAN: A Fully Parametric Survival Model Based on Kolmogorov-Arnold Networks
标题：SurvKAN：基于Kolmogorov-Arnold网络的全参数生存模型
链接：https://arxiv.org/abs/2602.02179

作者：Marina Mastroleo,Alberto Archetti,Federico Mastroleo,Matteo Matteucci

【13】Interpretable Tabular Foundation Models via In-Context Kernel Regression
标题：基于上下文核回归的可解释表格基础模型
链接：https://arxiv.org/abs/2602.02162

作者：Ratmir Miftachov,Bruno Charron,Simon Valentin

【14】Learning Generative Selection for Best-of-N
标题：学习N中最佳的生成性选择
链接：https://arxiv.org/abs/2602.02143

作者：Shubham Toshniwal,Aleksander Ficek,Siddhartha Jain,Wei Du,Vahid Noroozi,Sadegh Mahdavi,Somshubra Majumdar,Igor Gitman

【15】The Maximum von Neumann Entropy Principle: Theory and Applications in Machine Learning
标题：最大冯·诺伊曼熵原理：机器学习中的理论与应用
链接：https://arxiv.org/abs/2602.02117

作者：Youqi Wu,Farzan Farnia

【16】An Empirical Study of World Model Quantization
标题：世界模型量化的实证研究
链接：https://arxiv.org/abs/2602.02110

作者：Zhongqian Fu,Tianyi Zhao,Kai Han,Hang Zhou,Xinghao Chen,Yunhe Wang

【17】Learning Half-Spaces from Perturbed Contrastive Examples
标题：从受干扰的对比示例中学习半空间
链接：https://arxiv.org/abs/2602.02080

作者：Aryan Alavi Razavi Ravari,Farnam Mansouri,Yuxin Chen,Valentio Iverson,Adish Singla,Sandra Zilles

【18】Ultrafast On-chip Online Learning via Spline Locality in Kolmogorov-Arnold Networks
标题：Kolmogorov-Arnold网络中通过样条局部性的超快片上在线学习
链接：https://arxiv.org/abs/2602.02056

作者：Duc Hoang,Aarush Gupta,Philip Harris

【19】Hippasus: Effective and Efficient Automatic Feature Augmentation for Machine Learning Tasks on Relational Data
标题：Hippasus：关系数据上的机器学习任务的有效且高效的自动特征增强
链接：https://arxiv.org/abs/2602.02025

作者：Serafeim Papadias,Kostas Patroumpas,Dimitrios Skoutas
备注：13 pages, 7 figures, 9 tables

【20】Logic-Guided Vector Fields for Constrained Generative Modeling
标题：用于约束生成建模的逻辑引导向量场
链接：https://arxiv.org/abs/2602.02009

作者：Ali Baheri

【21】Optimizing Tensor Train Decomposition in DNNs for RISC-V Architectures Using Design Space Exploration and Compiler Optimizations
标题：使用设计空间探索和简化器优化优化RISC-V架构DNN中的张量序列分解
链接：https://arxiv.org/abs/2602.01996

作者：Theologos Anthimopoulos,Milad Kokhazadeh,Vasilios Kelefouras,Benjamin Himpel,Georgios Keramidas
备注：36 pages, 16 figures, this is the author-accepted version of the article published in ACM Transactions on Embedded Computing Systems (TECS), Vol. 24, No. 6

【22】SpikingGamma: Surrogate-Gradient Free and Temporally Precise Online Training of Spiking Neural Networks with Smoothed Delays
标题：SpikingGamma：具有平滑延迟的Spiking神经网络的无替代梯度且时间精确的在线训练
链接：https://arxiv.org/abs/2602.01978

作者：Roel Koopman,Sebastian Otte,Sander Bohté

【23】FlyPrompt: Brain-Inspired Random-Expanded Routing with Temporal-Ensemble Experts for General Continual Learning
标题：FlyPrompt：具有时间参与专家的脑启发随机扩展路由，用于一般持续学习
链接：https://arxiv.org/abs/2602.01976

作者：Hongwei Yan,Guanglong Sun,Kanglei Zhou,Qian Li,Liyuan Wang,Yi Zhong
备注：33 pages. Accepted by ICLR 2026

【24】Grounding Generated Videos in Feasible Plans via World Models
标题：通过世界模型将生成的视频作为可行计划的基础
链接：https://arxiv.org/abs/2602.01960

作者：Christos Ziakas,Amir Bar,Alessandra Russo

【25】Deep Multivariate Models with Parametric Conditionals
标题：带参数条件的深度多元模型
链接：https://arxiv.org/abs/2602.01953

作者：Dmitrij Schlesinger,Boris Flach,Alexander Shekhovtsov

【26】MACD: Model-Aware Contrastive Decoding via Counterfactual Data
标题：MACD：通过反事实数据的模型感知对比解码
链接：https://arxiv.org/abs/2602.01740

作者：Qixin Xiao,Kun Zhou

【27】On the Spatiotemporal Dynamics of Generalization in Neural Networks
标题：神经网络推广的时空动力学
链接：https://arxiv.org/abs/2602.01651

作者：Zichao Wei

【28】From Perception to Action: Spatial AI Agents and World Models
标题：从感知到行动：空间人工智能代理和世界模型
链接：https://arxiv.org/abs/2602.01644

作者：Gloria Felicia,Nolan Bryant,Handi Putra,Ayaan Gazali,Eliel Lobo,Esteban Rojas
备注：61 pages, 742 citations, 1 figure, 3 tables. Survey paper on spatial AI agents, embodied AI, graph neural networks, and world models

【29】What Do Agents Learn from Trajectory-SFT: Semantics or Interfaces?
标题：代理从轨迹SFT中学到什么：语义还是界面？
链接：https://arxiv.org/abs/2602.01611

作者：Weizheng Gu,Chengze Li,Zhuohao Yu,Mengyuan Sun,Zhibang Yang,Wei Wang,Hongrui Jia,Shikun Zhang,Wei Ye

【30】Universal Redundancies in Time Series Foundation Models
标题：时间序列基础模型中的普遍冗余
链接：https://arxiv.org/abs/2602.01605

作者：Anthony Bao,Venkata Hasith Vattikuti,Jeffrey Lai,William Gilpin

【31】Generative Visual Code Mobile World Models
标题：生成式视觉代码移动世界模型
链接：https://arxiv.org/abs/2602.01576

作者：Woosung Koh,Sungjun Han,Segyu Lee,Se-Young Yun,Jamin Shin
备注：Pre-print (technical report)

【32】Rod Flow: A Continuous-Time Model for Gradient Descent at the Edge of Stability
标题：棒流：稳定边缘梯度下降的连续时间模型
链接：https://arxiv.org/abs/2602.01480

作者：Eric Regis,Sinho Chewi

【33】Phase Transitions for Feature Learning in Neural Networks
标题：神经网络特征学习的阶段转变
链接：https://arxiv.org/abs/2602.01434

作者：Andrea Montanari,Zihao Wang
备注：74 pages; 17 pdf figures

【34】Nonlinear model reduction for transport-dominated problems
标题：运输主导问题的非线性模型约简
链接：https://arxiv.org/abs/2602.01397

作者：Jan S. Hesthaven,Benjamin Peherstorfer,Benjamin Unger

【35】High-accuracy sampling for diffusion models and log-concave distributions
标题：扩散模型和log-凹分布的高精度采样
链接：https://arxiv.org/abs/2602.01338

作者：Fan Chen,Sinho Chewi,Constantinos Daskalakis,Alexander Rakhlin

【36】PolySAE: Modeling Feature Interactions in Sparse Autoencoders via Polynomial Decoding
标题：PolyAE：通过多项解码建模稀疏自动编码器中的特征交互
链接：https://arxiv.org/abs/2602.01322

作者：Panagiotis Koromilas,Andreas D. Demou,James Oldfield,Yannis Panagakis,Mihalis Nicolaou

【37】Dispelling the Curse of Singularities in Neural Network Optimizations
标题：消除神经网络优化中的奇异性诅咒
链接：https://arxiv.org/abs/2602.01308

作者：Hengjie Cao,Mengyi Chen,Yifeng Yang,Fang Dong,Ruijun Huang,Anrui Chen,Jixian Zhou,Mingzhi Dong,Yujiang Wang,Dongsheng Li,Wenyi Fang,Yuanyi Lin,Fan Wu,Li Shang

【38】Gradient-Aligned Calibration for Post-Training Quantization of Diffusion Models
标题：扩散模型训练后量化的对象对齐校准
链接：https://arxiv.org/abs/2602.01289

作者：Dung Anh Hoang,Cuong Pham anh Trung Le,Jianfei Cai,Toan Do

【39】From Intents to Actions: Agentic AI in Autonomous Networks
标题：从意图到行动：自治网络中的大型人工智能
链接：https://arxiv.org/abs/2602.01271

作者：Burak Demirel,Pablo Soldati,Yu Wang

【40】Mechanistic Interpretability of Brain-to-Speech Models Across Speech Modes
标题：跨语音模式的脑到语音模型的机械解释性
链接：https://arxiv.org/abs/2602.01247

作者：Maryam Maghsoudi,Ayushi Mishra

【41】Learning from Anonymized and Incomplete Tabular Data
标题：从模拟和不完整的表格数据中学习
链接：https://arxiv.org/abs/2602.01217

作者：Lucas Lange,Adrian Böttinger,Victor Christen,Anushka Vidanage,Peter Christen,Erhard Rahm

【42】FedBGS: A Blockchain Approach to Segment Gossip Learning in Decentralized Systems
标题：FedBGS：一种在去中心化系统中细分八卦学习的区块链方法
链接：https://arxiv.org/abs/2602.01185

作者：Fabio Turazza,Marcello Pietri,Marco Picone,Marco Mamei
备注：Author-accepted manuscript of a paper published in the 2025 IEEE 45th International Conference on Distributed Computing Systems Workshops (ICDCSW), pp. 760-770, doi: 10.1109/ICDCSW63273.2025.00136

【43】Analyzing and Improving Diffusion Models for Time-Series Data Imputation: A Proximal Recursion Perspective
标题：分析和改进时间序列数据插补的扩散模型：近端回归的角度
链接：https://arxiv.org/abs/2602.01182

作者：Zhichao Chen,Hao Wang,Fangyikang Wang,Licheng Pan,Zhengnan Li,Yunfei Teng,Haoxuan Li,Zhouchen Lin

【44】A Unified Matrix-Spectral Framework for Stability and Interpretability in Deep Learning
标题：深度学习稳定性和可解释性的统一矩阵谱框架
链接：https://arxiv.org/abs/2602.01136

作者：Ronald Katende
备注：11 pages

【45】Long-range Modeling and Processing of Multimodal Event Sequences
标题：多峰事件序列的长期建模与处理
链接：https://arxiv.org/abs/2602.01125

作者：Jichu Li,Yilun Zhong,Zhiting Li,Feng Zhou,Quyu Kong

【46】On the Expressive Power of Permutation-Equivariant Weight-Space Networks
标题：论置换等变权空间网络的表现能力
链接：https://arxiv.org/abs/2602.01083

作者：Adir Dayan,Yam Eitan,Haggai Maron

【47】LASS-ODE: Scaling ODE Computations to Connect Foundation Models with Dynamical Physical Systems
标题：LASS-ODE：扩展ODE计算以连接基础模型与动态物理系统
链接：https://arxiv.org/abs/2602.01009

作者：Haoran Li,Chenhan Xiao,Lihao Mai,Yang Weng,Erik Blasch

【48】Multimodal Scientific Learning Beyond Diffusions and Flows
标题：超越扩散和流动的多模式科学学习
链接：https://arxiv.org/abs/2602.00960

作者：Leonardo Ferreira Guilhoto,Akshat Kaushal,Paris Perdikaris

【49】Over-Alignment vs Over-Fitting: The Role of Feature Learning Strength in Generalization
标题：过度对齐与过度匹配：特征学习强度在概括中的作用
链接：https://arxiv.org/abs/2602.00827

作者：Taesun Yeom,Taehyeok Ha,Jaeho Lee

【50】Learning in Bayesian Stackelberg Games With Unknown Follower's Types
标题：跟随者类型未知的Bayesian Stackelberg游戏中的学习
链接：https://arxiv.org/abs/2602.00771

作者：Matteo Bollini,Francesco Bacchiocchi,Samuel Coutts,Matteo Castiglioni,Alberto Marchesi

【51】Cross-Modal Binary Attention: An Energy-Efficient Fusion Framework for Audio-Visual Learning
标题：跨模态二元注意：一种能量有效的视听学习融合框架
链接：https://arxiv.org/abs/2602.00701

作者：Mohamed Saleh,Zahra Ahmadi

【52】Topology and Geometry of the Learning Space of ReLU Networks: Connectivity and Singularities
标题：ReLU网络学习空间的布局和几何：连通性和奇异性
链接：https://arxiv.org/abs/2602.00693

作者：Marco Nurisso,Pierrick Leroy,Giovanni Petri,Francesco Vaccarino
备注：Accepted to ICLR 2026. 32 pages, 13 figures

【53】Strong Linear Baselines Strike Back: Closed-Form Linear Models as Gaussian Process Conditional Density Estimators for TSAD
标题：强线性基线反击：封闭线性模型作为TSAD的高斯过程条件密度估计器
链接：https://arxiv.org/abs/2602.00672

作者：Aleksandr Yugay,Hang Cui,Changhua Pei,Alexey Zaytsev

【54】Equilibrium of Feasible Zone and Uncertain Model in Safe Exploration
标题：安全勘探中的可行区平衡与不确定模型
链接：https://arxiv.org/abs/2602.00636

作者：Yujie Yang,Zhilong Zheng,Shengbo Eben Li

【55】One Loss to Rule Them All: Marked Time-to-Event for Structured EHR Foundation Models
标题：一个失败的规则：结构化EHR基础模型的标记时间到事件
链接：https://arxiv.org/abs/2602.00541

作者：Zilin Jing,Vincent Jeanselme,Yuta Kobayashi,Simon A. Lee,Chao Pang,Aparajita Kashyap,Yanwei Li,Xinzhuo Jiang,Shalmali Joshi

【56】Invertible Memory Flow Networks
标题：可逆记忆流网络
链接：https://arxiv.org/abs/2602.00535

作者：Liyu Zerihun,Alexandr Plashchinsky

【57】AIRE-Prune: Asymptotic Impulse-Response Energy for State Pruning in State Space Models
标题：AIRE-Prune：状态空间模型中状态剪枝的渐近脉冲响应能量
链接：https://arxiv.org/abs/2602.00534

作者：Apurba Prasad Padhy,Fernando Camacho,Saibal Mukhopadhyay

【58】Contrastive Learning for Privacy Enhancements in Industrial Internet of Things
标题：工业物联网隐私增强的对比学习
链接：https://arxiv.org/abs/2602.00515

作者：Lin Liu,Rita Machacy,Simi Kuniyilh

【59】Parallel Stochastic Gradient-Based Planning for World Models
标题：并行随机基于对象的世界模型规划
链接：https://arxiv.org/abs/2602.00475

作者：Michael Psenka,Michael Rabbat,Aditi Krishnapriyan,Yann LeCun,Amir Bar
备注：23 pages, 7 figures

【60】Towards Building Non-Fine-Tunable Foundation Models
标题：迈向构建不可微调基础模型
链接：https://arxiv.org/abs/2602.00446

作者：Ziyao Wang,Nizhang Li,Pingzhi Li,Guoheng Sun,Tianlong Chen,Ang Li

【61】RePaint-Enhanced Conditional Diffusion Model for Parametric Engineering Designs under Performance and Parameter Constraints
标题：性能和参数约束下参数化工程设计的RePaint增强条件扩散模型
链接：https://arxiv.org/abs/2602.00384

作者：Ke Wang,Nguyen Gia Hien Vu,Yifan Tang,Mostafa Rahmani Dehaghani,G. Gary Wang

【62】Modeling Image-Caption Rating from Comparative Judgments
标题：根据比较判断建模图像字幕评级
链接：https://arxiv.org/abs/2602.00381

作者：Kezia Minni,Qiang Zhang,Monoshiz Mahbub Khan,Zhe Yu

【63】Agentic Framework for Epidemiological Modeling
标题：流行病学建模的统计框架
链接：https://arxiv.org/abs/2602.00299

作者：Rituparna Datta,Zihan Guan,Baltazar Espinoza,Yiqi Su,Priya Pitre,Srini Venkatramanan,Naren Ramakrishnan,Anil Vullikanti

【64】TABES: Trajectory-Aware Backward-on-Entropy Steering for Masked Diffusion Models
标题：TABEP：掩蔽扩散模型的轨迹感知逆向引导
链接：https://arxiv.org/abs/2602.00250

作者：Shreshth Saini,Avinab Saha,Balu Adsumilli,Neil Birkbeck,Yilin Wang,Alan C. Bovik

【65】Learning to Price: Interpretable Attribute-Level Models for Dynamic Markets
标题：学习定价：动态市场的可解释属性水平模型
链接：https://arxiv.org/abs/2602.00188

作者：Srividhya Sethuraman,Chandrashekar Lakshminarayanan
备注：Accepted in AAMAS 2026 - main track - full paper - 12 pages

【66】THDC: Training Hyperdimensional Computing Models with Backpropagation
标题：THDC：使用反向传播训练多维计算模型
链接：https://arxiv.org/abs/2602.00116

作者：Hanne Dejonghe,Sam Leroux
备注：Accepted to ESANN 2026

【67】Gauss-Newton Natural Gradient Descent for Shape Learning
标题：高斯-牛顿自然梯度下降用于形状学习
链接：https://arxiv.org/abs/2602.00099

作者：James King,Arturs Berzins,Siddhartha Mishra,Marius Zeinhofer
备注：16 Pages, 9 Figures, submitted to Computer-Aided Design

【68】Trade-offs Between Individual and Group Fairness in Machine Learning: A Comprehensive Review
标题：机器学习中个人和群体公平性之间的权衡：全面评论
链接：https://arxiv.org/abs/2602.00094

作者：Sandra Benítez-Peña,Blas Kolic,Victoria Menendez,Belén Pulido

【69】Interpreting and Controlling Model Behavior via Constitutions for Atomic Concept Edits
标题：通过原子概念编辑宪法解释和控制模型行为
链接：https://arxiv.org/abs/2602.00092

作者：Neha Kalibhat,Zi Wang,Prasoon Bajpai,Drew Proud,Wenjun Zeng,Been Kim,Mani Malek

【70】Repair Brain Damage: Real-Numbered Error Correction Code for Neural Network
标题：修复脑损伤：神经网络的实编号错误纠正代码
链接：https://arxiv.org/abs/2602.00076

作者：Ziqing Li,Myung Cho,Qiutong Jin,Weiyu Xu
备注：6 pages, 3 figures

【71】SCPL: Enhancing Neural Network Training Throughput with Decoupled Local Losses and Model Parallelism
标题：SIPL：通过脱钩局部损失和模型并行主义来增强神经网络训练输出
链接：https://arxiv.org/abs/2602.00062

作者：Ming-Yao Ho,Cheng-Kai Wang,You-Teng Lin,Hung-Hsuan Chen

【72】Lightweight Edge Learning via Dataset Pruning
标题：通过数据集修剪的轻量级边缘学习
链接：https://arxiv.org/abs/2602.00047

作者：Laha Ale,Hu Luo,Mingsheng Cao,Shichao Li,Huanlai Xing,Haifeng Sun
备注：11 pages, 10 figures

【73】Construct, Align, and Reason: Large Ontology Models for Enterprise Knowledge Management
标题：构建、对齐和推理：企业知识管理的大型实体模型
链接：https://arxiv.org/abs/2602.00029

作者：Yao Zhang,Hongyin Zhu

【74】Measurement for Opaque Systems: Multi-source Triangulation with Interpretable Machine Learning
标题：不透明系统的测量：具有可解释机器学习的多源三角测量
链接：https://arxiv.org/abs/2602.00022

作者：Margaret Foster
备注：16 pages, 6 figures, 3 tables, 9-page appendix

【75】Linear-PAL: A Lightweight Ranker for Mitigating Shortcut Learning in Personalized, High-Bias Tabular Ranking
标题：Linear-PAL：一种轻量级排名，用于缓解个性化、高偏差表格排名中的工作量学习
链接：https://arxiv.org/abs/2602.00013

作者：Vipul Dinesh Pawar

【76】Assessing the Impact of Image Dataset Features on Privacy-Preserving Machine Learning
标题：评估图像数据集特征对隐私保护机器学习的影响
链接：https://arxiv.org/abs/2409.01329

作者：Lucas Lange,Maurice-Maximilian Heykeroth,Erhard Rahm
备注：Accepted at 21st Conference on Database Systems for Business, Technology and Web (BTW 2025)

【77】Full-Batch Gradient Descent Outperforms One-Pass SGD: Sample Complexity Separation in Single-Index Learning
标题：全批梯度下降优于一遍新元：单指标学习中的样本复杂性分离
链接：https://arxiv.org/abs/2602.02431

作者：Filip Kovačević,Hong Chang Ji,Denny Wu,Mahdi Soltanolkotabi,Marco Mondelli

【78】Learning Beyond the Gaussian Data: Learning Dynamics of Neural Networks on an Expressive and Cumulant-Controllable Data Model
标题：超越高斯数据的学习：基于可表达和累积可控数据模型的神经网络学习动力学
链接：https://arxiv.org/abs/2602.02153

作者：Onat Ure,Samet Demir,Zafer Dogan
备注：ICASSP 2026, 5 pages, 2 figures

【79】FluxNet: Learning Capacity-Constrained Local Transport Operators for Conservative and Bounded PDE Surrogates
标题：FluxNet：保守和有限的PED代理人的学习能力受限的当地交通运营商
链接：https://arxiv.org/abs/2602.01941

作者：Zishuo Lan,Junjie Li,Lei Wang,Jincheng Wang

【80】Learning Sequential Decisions from Multiple Sources via Group-Robust Markov Decision Processes
标题：通过群体鲁棒马尔科夫决策过程学习来自多个来源的顺序决策
链接：https://arxiv.org/abs/2602.01825

作者：Mingyuan Xu,Zongqi Xia,Tianxi Cai,Doudou Zhou,Nian Si

【81】Density-Informed Pseudo-Counts for Calibrated Evidential Deep Learning
标题：用于校准证据深度学习的密度知情伪计数
链接：https://arxiv.org/abs/2602.01477

作者：Pietro Carlotti,Nevena Gligić,Arya Farahi

【82】WAKESET: A Large-Scale, High-Reynolds Number Flow Dataset for Machine Learning of Turbulent Wake Dynamics
标题：Wakeset：用于湍流尾流动力学机器学习的大规模、高雷诺数流量数据集
链接：https://arxiv.org/abs/2602.01379

作者：Zachary Cooper-Baldock,Paulo E. Santos,Russell S. A. Brinkworth,Karl Sammut
备注：27 pages, 7 figures, 2 tables

【83】Equivalence of Privacy and Stability with Generalization Guarantees in Quantum Learning
标题：量子学习中隐私性和稳定性与一般化保证的等效性
链接：https://arxiv.org/abs/2602.01177

作者：Ayanava Dasgupta,Naqueeb Ahmad Warsi,Masahito Hayashi
备注：22 pages, 3 figures; This paper studies the interplay between privacy, stability, and generalization in quantum learning. The stability analysis in Section IV overlaps with the companion preprint arXiv:2511.01467, which focuses on quantum information ordering, while the present work focuses on generalization guarantees in quantum learning

【84】Robust Machine Learning Framework for Reliable Discovery of High-Performance Half-Heusler Thermoelectrics
标题：稳健的机器学习框架，用于可靠地发现高性能半赫斯勒热电器
链接：https://arxiv.org/abs/2602.01149

作者：Shoeb Athar,Adrien Mecibah,Philippe Jund
备注：44 pages, 8 figures. Submitted for publication

【85】The Quantum Learning Menagerie (A survey on Quantum learning for Classical concepts)
标题：量子学习动物园（经典概念量子学习的调查）
链接：https://arxiv.org/abs/2602.01054

作者：Sagnik Chatterjee

【86】Emergence of Distortions in High-Dimensional Guided Diffusion Models
标题：多维引导扩散模型中失真的出现
链接：https://arxiv.org/abs/2602.00716

作者：Enrico Ventura,Beatrice Achilli,Luca Ambrogioni,Carlo Lucibello
备注：ICML 2026 submission, 29 pages, 16 figures

【87】Multimodal Machine Learning for Integrating Heterogeneous Analytical Systems
标题：用于集成异构分析系统的多模态机器学习
链接：https://arxiv.org/abs/2602.00590

作者：Shun Muroga,Hideaki Nakajima,Taiyo Shimizu,Kazufumi Kobashi,Kenji Hata
备注：12 pages, 4 figures, 2 tables

【88】Singular Bayesian Neural Networks
标题：奇异Bayesian神经网络
链接：https://arxiv.org/abs/2602.00387

作者：Mame Diarra Toure,David A. Stephens
备注：8 pages Main text, 53 pages Appendix, 20 figures

【89】Parametrization of subgrid scales in long-term simulations of the shallow-water equations using machine learning and convex limiting
标题：使用机器学习和凸限制进行浅水方程组长期模拟中的次网格尺度参数化
链接：https://arxiv.org/abs/2602.00378

作者：Md Amran Hossan Mojamder,Zhihang Xu,Min Wang,Ilya Timofeyev

【90】Quantum Circuit-Based Learning Models: Bridging Quantum Computing and Machine Learning
标题：基于量子电路的学习模型：桥接量子计算和机器学习
链接：https://arxiv.org/abs/2602.00048

作者：Fan Fan,Yilei Shi,Mihai Datcu,Bertrand Le Saux,Luigi Iapichino,Francesca Bovolo,Silvia Liberata Ullo,Xiao Xiang Zhu

其他(126篇)

【1】Reward-free Alignment for Conflicting Objectives
标题：预算目标的免奖励调整
链接：https://arxiv.org/abs/2602.02495

作者：Peter Chen,Xiaopeng Li,Xi Chen,Tianyi Lin
备注：27 pages

【2】MEG-XL: Data-Efficient Brain-to-Text via Long-Context Pre-Training
标题：MEG-XL：通过长上下文预训练实现数据高效的大脑到文本
链接：https://arxiv.org/abs/2602.02494

作者：Dulhan Jayalath,Oiwi Parker Jones
备注：19 pages, 8 figures, 5 tables

【3】HumanX: Toward Agile and Generalizable Humanoid Interaction Skills from Human Videos
标题：HumanX：从人类视频中迈向敏捷和可推广的人形交互技能
链接：https://arxiv.org/abs/2602.02473

作者：Yinhuai Wang,Qihan Zhao,Yuen Fui Lau,Runyi Yu,Hok Wai Tsui,Qifeng Chen,Jingbo Wang,Jiangmiao Pang,Ping Tan

【4】Finite-Sample Wasserstein Error Bounds and Concentration Inequalities for Nonlinear Stochastic Approximation
标题：非线性随机逼近的样本Wasserstein误差界和集中度不等式
链接：https://arxiv.org/abs/2602.02445

作者：Seo Taek Kong,R. Srikant

【5】Certain Head, Uncertain Tail: Expert-Sample for Test-Time Scaling in Fine-Grained MoE
标题：一定的头，不确定的尾：细粒度MoE测试时间缩放的专家样本
链接：https://arxiv.org/abs/2602.02443

作者：Yuanteng Chen,Peisong Wang,Nanxin Zeng,Yuantian Shao,Gang Li,Jing Liu,Jian Cheng
备注：24 pages, 13 figures

【6】Poly-attention: a general scheme for higher-order self-attention
标题：多重注意力：高级自我注意力的一般方案
链接：https://arxiv.org/abs/2602.02422

作者：Sayak Chakrabarti,Toniann Pitassi,Josh Alman

【7】Misconception Diagnosis From Student-Tutor Dialogue: Generate, Retrieve, Rerank
标题：学生与导师对话的误解诊断：生成、删除、重新排序
链接：https://arxiv.org/abs/2602.02414

作者：Joshua Mitton,Prarthana Bhattacharyya,Digory Smith,Thomas Christie,Ralph Abboud,Simon Woodhead
备注：21 pages, 8 figures, 8 tables. Joshua Mitton and Prarthana Bhattacharyya contributed equally to this paper

【8】Masked Autoencoders as Universal Speech Enhancer
标题：屏蔽自动编码器作为通用语音增强器
链接：https://arxiv.org/abs/2602.02413

作者：Rajalaxmi Rajagopalan,Ritwik Giri,Zhiqiang Tang,Kyu Han

【9】Live-Evo: Online Evolution of Agentic Memory from Continuous Feedback
标题：Live-Evo：持续反馈中记忆的在线进化
链接：https://arxiv.org/abs/2602.02369

作者：Yaolun Zhang,Yiran Wu,Yijiong Yu,Qingyun Wu,Huazheng Wang
备注：13 pages

【10】DFKI-Speech System for WildSpoof Challenge: A robust framework for SASV In-the-Wild
标题：用于WildSpoof挑战的DFKI语音系统：SASV野外的强大框架
链接：https://arxiv.org/abs/2602.02286

作者：Arnab Das,Yassine El Kheir,Enes Erdem Erdogan,Feidi Kallel,Tim Polzehl,Sebastian Moeller

【11】Backpropagation as Physical Relaxation: Exact Gradients in Finite Time
标题：作为物理放松的反向传播：有限时间内的精确结果
链接：https://arxiv.org/abs/2602.02281

作者：Antonino Emanuele Scurria
备注：15 pages, 8 figures

【12】Unlocking the Duality between Flow and Field Matching
标题：解锁流场匹配之间的二元性
链接：https://arxiv.org/abs/2602.02261

作者：Daniil Shlenskii,Alexander Varlamov,Nazar Buzun,Alexander Korotin

【13】Spectral Superposition: A Theory of Feature Geometry
标题：谱叠加：特征几何理论
链接：https://arxiv.org/abs/2602.02224

作者：Georgi Ivanov,Narmeen Oozeer,Shivam Raval,Tasana Pejovic,Shriyash Upadhyay,Amir Abdullah

【14】Scientific Theory of a Black-Box: A Life Cycle-Scale XAI Framework Based on Constructive Empiricism
标题：黑匣子的科学理论：基于建构性经验主义的生命周期规模XAI框架
链接：https://arxiv.org/abs/2602.02215

作者：Sebastian Müller,Vanessa Toborek,Eike Stadtländer,Tamás Horváth,Brendan Balcerak Jackson,Christian Bauckhage

【15】Efficient Neural Controlled Differential Equations via Attentive Kernel Smoothing
标题：通过专注的核平滑实现高效的神经控制方程
链接：https://arxiv.org/abs/2602.02157

作者：Egor Serov,Ilya Kuleshov,Alexey Zaytsev

【16】EvoMU: Evolutionary Machine Unlearning
标题：EvoMU：进化机器遗忘
链接：https://arxiv.org/abs/2602.02139

作者：Pawel Batorski,Paul Swoboda

【17】Scalable Spatio-Temporal SE(3) Diffusion for Long-Horizon Protein Dynamics
标题：长视野蛋白质动力学的可扩展时空SE（3）扩散
链接：https://arxiv.org/abs/2602.02128

作者：Nima Shoghi,Yuxuan Liu,Yuning Shen,Rob Brekelmans,Pan Li,Quanquan Gu
备注：For associated project page, see https://bytedance-seed.github.io/ConfRover/starmd

【18】Efficient Swap Regret Minimization in Combinatorial Bandits
标题：组合盗贼中的有效交换后悔最小化
链接：https://arxiv.org/abs/2602.02087

作者：Andreas Kontogiannis,Vasilis Pollatos,Panayotis Mertikopoulos,Ioannis Panageas
备注：Accepted at AISTATS 2026

【19】BAPS: A Fine-Grained Low-Precision Scheme for Softmax in Attention via Block-Aware Precision reScaling
标题：BAPS：通过块感知精度重新缩放，Softmax的细粒度低精度方案受到关注
链接：https://arxiv.org/abs/2602.02071

作者：Zisheng Ye, Xiaoyu He, Maoyuan Song, Guoliang Qiu, Chao Liao, Chen Wu, Yonggang Sun, Zhichun Li, Xiaoru Xie, Yuanyong Luo, Hu Liu, Pinyan Lu, Heng Liao

【20】FiLoRA: Focus-and-Ignore LoRA for Controllable Feature Reliance
标题：FiLoRA：聚焦并忽略LoRA以实现可控功能依赖
链接：https://arxiv.org/abs/2602.02060

作者：Hyunsuk Chung,Caren Han,Yerin Choi,Seungyeon Ji,Jinwoo Kim,Eun-Jung Holden,Kyungreem Han

【21】On Stability and Robustness of Diffusion Posterior Sampling for Bayesian Inverse Problems
标题：Bayesian逆问题扩散后验抽样的稳定性和鲁棒性
链接：https://arxiv.org/abs/2602.02045

作者：Yiming Yang,Xiaoyuan Cheng,Yi He,Kaiyu Li,Wenxuan Yuan,Zhuo Sun

【22】Bandwidth-Efficient Multi-Agent Communication through Information Bottleneck and Vector Quantization
标题：通过信息瓶颈和载体量化实现带宽高效的多智能体通信
链接：https://arxiv.org/abs/2602.02035

作者：Ahmad Farooq,Kamran Iqbal
备注：Accepted at the 2026 IEEE International Conference on Robotics and Automation (ICRA 2026), Vienna, Austria. 9 pages, 4 figures, 6 tables

【23】Scale-covariant spiking wavelets
标题：尺度协变尖峰子波
链接：https://arxiv.org/abs/2602.02020

作者：Jens Egholm Pedersen,Tony Lindeberg,Peter Gerstoft

【24】DASH: Faster Shampoo via Batched Block Preconditioning and Efficient Inverse-Root Solvers
标题：DASH：通过批量块预处理和高效的反根解决方案实现更快的洗发水
链接：https://arxiv.org/abs/2602.02016

作者：Ionut-Vlad Modoranu,Philip Zmushko,Erik Schultheis,Mher Safaryan,Dan Alistarh

【25】Robust Domain Generalization under Divergent Marginal and Conditional Distributions
标题：分歧边缘和条件分布下的鲁棒领域推广
链接：https://arxiv.org/abs/2602.02015

作者：Jewon Yeom,Kyubyung Chae,Hyunggyu Lim,Yoonna Oh,Dongyoon Yang,Taesup Kim

【26】SNAP: A Self-Consistent Agreement Principle with Application to Robust Computation
标题：SNAP：自相容一致原则及其在鲁棒计算中的应用
链接：https://arxiv.org/abs/2602.02013

作者：Xiaoyi Jiang,Andreas Nienkötter

【27】Position: The Need for Ultrafast Training
标题：职位：超快训练的需要
链接：https://arxiv.org/abs/2602.02005

作者：Duc Hoang
备注：Position paper at the 2nd Workshop on Domain-Specialized FPGAs (WDSFPGA 2026)

【28】SAME: Stabilized Mixture-of-Experts for Multimodal Continual Instruction Tuning
标题：：用于多模式连续指令调优的稳定专家混合
链接：https://arxiv.org/abs/2602.01990

作者：Zhen-Hao Xie,Jun-Tao Tang,Yu-Cheng Shi,Han-Jia Ye,De-Chuan Zhan,Da-Wei Zhou

【29】Self-Consolidation for Self-Evolving Agents
标题：自我进化代理的自我整合
链接：https://arxiv.org/abs/2602.01966

作者：Hongzhuo Yu,Fei Zhu,Guo-Sen Xie,Ling Shao

【30】VLM-Guided Experience Replay
标题：VLM引导的体验重播
链接：https://arxiv.org/abs/2602.01915

作者：Elad Sharony,Tom Jurgenson,Orr Krupnik,Dotan Di Castro,Shie Mannor

【31】Data- and Variance-dependent Regret Bounds for Online Tabular MDPs
标题：在线表格式MDP的数据和方差相关的遗憾界限
链接：https://arxiv.org/abs/2602.01903

作者：Mingyi Li,Taira Tsuchiya,Kenji Yamanishi
备注：80pages, 4tables

【32】IRIS: Implicit Reward-Guided Internal Sifting for Mitigating Multimodal Hallucination
标题：IRIS：隐性奖励引导的内部筛选，缓解多模式幻觉
链接：https://arxiv.org/abs/2602.01769

作者：Yuanshuai Li,Yuping Yan,Jirui Han,Fei Ming,Lingjuan Lv,Yaochu Jin

【33】A Provable Expressiveness Hierarchy in Hybrid Linear-Full Attention
标题：混合线性中可证明的表达层次-充分注意
链接：https://arxiv.org/abs/2602.01763

作者：Xiaowei Ye,Xiaoyu He,Chao Liao,Chen Wu,Pinyan Lu

【34】Controlling Exploration-Exploitation in GFlowNets via Markov Chain Perspectives
标题：通过马尔科夫链视角控制GFlowNets中的探索利用
链接：https://arxiv.org/abs/2602.01749

作者：Lin Chen,Samuel Drapeau,Fanghao Shao,Xuekai Zhu,Bo Xue,Yunchong Song,Mathieu Laurière,Zhouhan Lin

【35】Enhancing Automated Essay Scoring with Three Techniques: Two-Stage Fine-Tuning, Score Alignment, and Self-Training
标题：通过三种技术增强自动论文评分：两阶段微调、分数对齐和自我训练
链接：https://arxiv.org/abs/2602.01747

作者：Hongseok Choi,Serynn Kim,Wencke Liermann,Jin Seong,Jin-Xia Huang
备注：22 pages, 4 figures

【36】Softmax Linear Attention: Reclaiming Global Competition
标题：Softmax线性关注：夺回全球竞争
链接：https://arxiv.org/abs/2602.01744

作者：Mingwei Xu,Xuan Lin,Xinnan Guo,Wanqing Xu,Wanyun Cui
备注：11 pages,4 figures

【37】Revisiting Generalization Measures Beyond IID: An Empirical Study under Distributional Shift
标题：重新审视IID之外的概括措施：分配转变下的实证研究
链接：https://arxiv.org/abs/2602.01718

作者：Sora Nakai,Youssef Fadhloun,Kacem Mathlouthi,Kotaro Yoshida,Ganesh Talluri,Ioannis Mitliagkas,Hiroki Naganuma

【38】Towards Autonomous Instrument Tray Assembly for Sterile Processing Applications
标题：面向无菌处理应用的自主器械托盘组件
链接：https://arxiv.org/abs/2602.01679

作者：Raghavasimhan Sankaranarayanan,Paul Stuart,Nicholas Ahn,Arno Sungarian,Yash Chitalia
备注：7 pages, 9 figures, 2026 International Symposium on Medical Robotics

【39】TRIP-Bench: A Benchmark for Long-Horizon Interactive Agents in Real-World Scenarios
标题：TRIP-Bench：现实世界场景中长视野互动代理的基准
链接：https://arxiv.org/abs/2602.01675

作者：Yuanzhe Shen,Zisu Huang,Zhengyuan Wang,Muzhao Tian,Zhengkang Guo,Chenyang Zhang,Shuaiyu Zhou,Zengjie Hu,Dailin Li,Jingwen Xu,Kaimin Wang,Wenhao Liu,Tianlong Li,Fengpeng Yue,Feng Hong,Cao Liu,Ke Zeng
备注：40 pages, 6figures

【40】The Effect of Mini-Batch Noise on the Implicit Bias of Adam
标题：小批量噪音对亚当内隐偏见的影响
链接：https://arxiv.org/abs/2602.01642

作者：Matias D. Cattaneo,Boris Shigida

【41】The Multiple Ticket Hypothesis: Random Sparse Subnetworks Suffice for RLVR
标题：多票假设：随机稀疏子网络足以满足WLVR
链接：https://arxiv.org/abs/2602.01599

作者：Israel Adewuyi,Solomon Okibe,Vladmir Ivanov

【42】Genus-0 Surface Parameterization using Spherical Beltrami Differentials
标题：使用球形Beltrami差异的Genus-0表面参数化
链接：https://arxiv.org/abs/2602.01589

作者：Zhehao Xu,Lok Ming Lui

【43】On the Fragility of AI-Based Channel Decoders under Small Channel Perturbations
标题：小通道扰动下基于人工智能的通道解码器的脆弱性
链接：https://arxiv.org/abs/2602.01582

作者：Haoyu Lei,Mohammad Jalali,Chin Wa Lau,Farzan Farnia

【44】Local Exponential Stability of Mean-Field Langevin Descent-Ascent in Wasserstein Space
标题：Wasserstein空间中平均场Langevin下降-上升的局部指数稳定性
链接：https://arxiv.org/abs/2602.01564

作者：Geuntaek Seo,Minseop Shin,Pierre Monmarché,Beomjun Choi

【45】InfoTok: Regulating Information Flow for Capacity-Constrained Shared Visual Tokenization in Unified MLLMs
标题：InfoTok：在统一MLLM中规范容量限制的共享视觉代币化的信息流
链接：https://arxiv.org/abs/2602.01554

作者：Lv Tang,Tianyi Zheng,Bo Li,Xingyu Li

【46】When Is Rank-1 Enough? Geometry-Guided Initialization for Parameter-Efficient Fine-Tuning
标题：一级什么时候够？几何引导的子菜单，实现参数高效的微调
链接：https://arxiv.org/abs/2602.01522

作者：Haoran Zhao,Soyeon Caren Han,Eduard Hovy

【47】Causal Preference Elicitation
标题：因果偏好启发
链接：https://arxiv.org/abs/2602.01483

作者：Edwin V. Bonilla,He Zhao,Daniel M. Steinberg

【48】P-EAGLE: Parallel-Drafting EAGLE with Scalable Training
标题：P-EAGLE：具有可扩展训练的并行起草EAGLE
链接：https://arxiv.org/abs/2602.01469

作者：Mude Hui,Xin Huang,Jaime Campos Salas,Yue Sun,Nathan Pemberton,Xiang Song,Ashish Khetan,George Karypis

【49】A Statistical Theory of Gated Attention through the Lens of Hierarchical Mixture of Experts
标题：从分层专家混合角度看门控注意力的统计理论
链接：https://arxiv.org/abs/2602.01468

作者：Viet Nguyen,Tuan Minh Pham,Thinh Cao,Tan Dinh,Huy Nguyen,Nhat Ho,Alessandro Rinaldo
备注：Viet Nguyen, Tuan Minh Pham, and Thinh Cao contributed equally to this work

【50】Provable Cooperative Multi-Agent Exploration for Reward-Free MDPs
标题：无奖励MDPs的可证明合作多代理探索
链接：https://arxiv.org/abs/2602.01453

作者：Idan Barnea,Orin Levy,Yishay Mansour

【51】The Gradient-Causal Gap: Why Gradient Importance Fails on Complex Tasks
标题：因果差距：为什么梯度重要性在复杂任务中失败
链接：https://arxiv.org/abs/2602.01442

作者：Donald Ye
备注：8 pages, 4 figures. Submitted to the ICLR 2026 Workshop on Latent & Implicit Thinking (LIT). Code:https://anonymous.4open.science/r/ICLR_2026_LIT-workshop_CG-D42B

【52】DCD: Decomposition-based Causal Discovery from Autocorrelated and Non-Stationary Temporal Data
标题：DID：从自相关和非平稳时间数据中基于分解的因果发现
链接：https://arxiv.org/abs/2602.01433

作者：Muhammad Hasan Ferdous,Md Osman Gani

【53】Building Better Deception Probes Using Targeted Instruction Pairs
标题：使用目标指令对构建更好的欺骗探针
链接：https://arxiv.org/abs/2602.01425

作者：Vikram Natarajan,Devina Jain,Shivam Arora,Satvik Golechha,Joseph Bloom

【54】PolyGen: Fully Synthetic Vision-Language Training via Multi-Generator Ensembles
标题：PolyGen：通过多生成器集成进行完全合成的视觉语言训练
链接：https://arxiv.org/abs/2602.01370

作者：Leonardo Brusini,Cristian Sbrolli,Eugenio Lomurno,Toshihiko Yamasaki,Matteo Matteucci

【55】Imperfect Influence, Preserved Rankings: A Theory of TRAK for Data Attribution
标题：不完美的影响力，保留的排名：TRAK数据归因理论
链接：https://arxiv.org/abs/2602.01312

作者：Han Tong,Shubhangi Ghosh,Haolin Zou,Arian Maleki

【56】The BoBW Algorithms for Heavy-Tailed MDPs
标题：用于重尾MDP的BoBW算法
链接：https://arxiv.org/abs/2602.01295

作者：Yu Chen,Yuhao Liu,Jiatai Huang,Yihan Du,Longbo Huang

【57】Richer Bayesian Last Layers with Subsampled NTK Features
标题：具有二次抽样NTK特征的更丰富的Bayesian最后层
链接：https://arxiv.org/abs/2602.01279

作者：Sergio Calvo-Ordoñez,Jonathan Plenk,Richard Bergna,Álvaro Cartea,Yarin Gal,Jose Miguel Hernández-Lobato,Kamil Ciosek
备注：Preprint, work in progress

【58】Diving into Kronecker Adapters: Component Design Matters
标题：深入研究克罗纳克适配器：组件设计很重要
链接：https://arxiv.org/abs/2602.01267

作者：Jiayu Bai,Danchen Yu,Zhenyu Liao,TianQi Hou,Feng Zhou,Robert C. Qiu,Zenan Ling

【59】MiTA Attention: Efficient Fast-Weight Scaling via a Mixture of Top-$k$ Activations
标题：MiTA注意力：通过混合顶级$k$激活进行高效快速减肥
链接：https://arxiv.org/abs/2602.01219

作者：Qishuai Wen,Zhiyuan Huang,Xianghan Meng,Wei He,Chun-Guang Li

【60】SimpleGPT: Improving GPT via A Simple Normalization Strategy
标题：SimpleGPT：通过简单的规范化策略改善GPT
链接：https://arxiv.org/abs/2602.01212

作者：Marco Chen,Xianbiao Qi,Yelin He,Jiaquan Ye,Rong Xiao
备注：We propose SimpleGPT, a simple yet effective GPT model, and provide theoretical insights into its mathematical foundations. We validate our theoretical findings through extensive experiments on large GPT models at parameter scales 1B, 1.4B, 7B and 8B

【61】Attention Sink Forges Native MoE in Attention Layers: Sink-Aware Training to Address Head Collapse
标题：注意力下沉在注意力层中锻造了本土MoE：解决头部塌陷问题的下沉意识训练
链接：https://arxiv.org/abs/2602.01203

作者：Zizhuo Fu,Wenxuan Zeng,Runsheng Wang,Meng Li

【62】Unraveling the Hidden Dynamical Structure in Recurrent Neural Policies
标题：揭开循环神经策略中隐藏的动态结构
链接：https://arxiv.org/abs/2602.01196

作者：Jin Li,Yue Wu,Mengsha Huang,Yuhao Sun,Hao He,Xianyuan Zhan

【63】Refining Context-Entangled Content Segmentation via Curriculum Selection and Anti-Curriculum Promotion
标题：通过课程选择和反课程推广完善上下文相关内容细分
链接：https://arxiv.org/abs/2602.01183

作者：Chunming He,Rihan Zhang,Fengyang Xiao,Dingming Zhang,Zhiwen Cao,Sina Farsiu
备注：8 figures, 11 tables

【64】Capabilities and Fundamental Limits of Latent Chain-of-Thought
标题：潜在思想链的能力和基本限制
链接：https://arxiv.org/abs/2602.01148

作者：Jiaxuan Zou,Yaozhong Xiong,Yong Liu

【65】Generalized Radius and Integrated Codebook Transforms for Differentiable Vector Quantization
标题：可微量化的广义半径和集成码本变换
链接：https://arxiv.org/abs/2602.01140

作者：Haochen You,Heng Zhang,Hongyang He,Yuqi Li,Baojing Liu
备注：This paper has been accepted as a conference paper at CPAL 2026

【66】OLion: Approaching the Hadamard Ideal by Intersecting Spectral and $\ell_{\infty}$ Implicit Biases
链接：https://arxiv.org/abs/2602.01105

作者：Zixiao Wang,Yifei Shen,Huishuai Zhang
备注：23 pages

【67】Superposition unifies power-law training dynamics
标题：叠加统一了权力定律训练动态
链接：https://arxiv.org/abs/2602.01045

作者：Zixin Jessie Chen,Hao Chen,Yizhou Liu,Jeff Gore
备注：17 pages, 14 figures

【68】On the Spectral Flattening of Quantized Embeddings
标题：关于量化嵌入的谱平坦化
链接：https://arxiv.org/abs/2602.00969

作者：Junlin Huang,Wenyi Fang,Zhenheng Tang,Yuxin Wang,Xueze Kang,Yang Zheng,Bo Li,Xiaowen Chu

【69】SAGE: Agentic Framework for Interpretable and Clinically Translatable Computational Pathology Biomarker Discovery
标题：SAGE：可解释和临床可翻译的计算病理学生物标志物发现的抽象框架
链接：https://arxiv.org/abs/2602.00953

作者：Sahar Almahfouz Nasser,Juan Francisco Pesantez Borja,Jincheng Liu,Tanvir Hasan,Zenghan Wang,Suman Ghosh,Sandeep Manandhar,Shikhar Shiromani,Twisha Shah,Naoto Tokuyama,Anant Madabhushi

【70】RoDiF: Robust Direct Fine-Tuning of Diffusion Policies with Corrupted Human Feedback
标题：RoDiF：对人类反馈受损的扩散政策进行稳健的直接微调
链接：https://arxiv.org/abs/2602.00886

作者：Amitesh Vatsa,Zhixian Xie,Wanxin Jin

【71】Test-time Generalization for Physics through Neural Operator Splitting
标题：通过神经运算符分裂进行物理测试时推广
链接：https://arxiv.org/abs/2602.00884

作者：Louis Serrano,Jiequn Han,Edouard Oyallon,Shirley Ho,Rudy Morel

【72】Improving Flow Matching by Aligning Flow Divergence
标题：通过调整流量分歧改善流量匹配
链接：https://arxiv.org/abs/2602.00869

作者：Yuhao Huang,Taos Transue,Shih-Hsin Wang,William Feldman,Hong Zhang,Bao Wang
备注：Published in ICML 2025

【73】Multi-Head Attention Is a Multi-Player Game
标题：多头注意力是一个多玩家游戏
链接：https://arxiv.org/abs/2602.00861

作者：Kushal Chakrabarti,Nirmal Balachundar

【74】JTok: On Token Embedding as another Axis of Scaling Law via Joint Token Self-modulation
标题：JTok：通过联合代币自调制将代币嵌入作为缩放定律的另一个轴
链接：https://arxiv.org/abs/2602.00800

作者：Yebin Yang,Huaijin Wu,Fu Guo,Lin Yao,Xiaohan Qin,Jingzhi Wang,Debing Zhang,Junchi Yan

【75】Latent Shadows: The Gaussian-Discrete Duality in Masked Diffusion
标题：潜伏的阴影：掩蔽扩散中的高斯-离散二元论
链接：https://arxiv.org/abs/2602.00792

作者：Guinan Chen,Xunpeng Huang,Ying Sun,Shijin Wang,Yanyong Zhang,Chao Wang
备注：10 pages

【76】Fast Non-Episodic Finite-Horizon RL with K-Step Lookahead Thresholding
标题：快速非剧集性幻想地平线RL，带K-Step前瞻性预设
链接：https://arxiv.org/abs/2602.00781

作者：Jiamin Xu,Kyra Gan

【77】BLOCK-EM: Preventing Emergent Misalignment by Blocking Causal Features
标题：BLOCK-EM：通过阻止因果特征来防止紧急失调
链接：https://arxiv.org/abs/2602.00767

作者：Muhammed Ustaomeroglu,Guannan Qu
备注：41 pages, 32 figures. Code available

【78】Rethinking Hallucinations: Correctness, Consistency, and Prompt Multiplicity
标题：重新思考幻觉：正确性、一致性和迅速的多重性
链接：https://arxiv.org/abs/2602.00723

作者：Prakhar Ganesh,Reza Shokri,Golnoosh Farnadi
备注：To appear at EACL 2026

【79】Scalable Generative Game Engine: Breaking the Resolution Wall via Hardware-Algorithm Co-Design
标题：可扩展生成游戏引擎：通过硬件算法协同设计打破分辨率墙
链接：https://arxiv.org/abs/2602.00608

作者：Wei Zeng,Xuchen Li,Ruili Feng,Zhen Liu,Fengwei An,Jian Zhao
备注：Preprint, Under Review

【80】Actor-Dual-Critic Dynamics for Zero-sum and Identical-Interest Stochastic Games
标题：零和与同利随机博弈的行为者二元批评动力学
链接：https://arxiv.org/abs/2602.00606

作者：Ahmed Said Donmez,Yuksel Arslantas,Muhammed O. Sayin

【81】Safe Langevin Soft Actor Critic
标题：安全朗之万软演员评论家
链接：https://arxiv.org/abs/2602.00587

作者：Mahesh Keswani,Samyak Jain,Raunak P. Bhattacharyya
备注：20 pages, 12 figures

【82】Beyond the Node: Clade-level Selection for Efficient MCTS in Automatic Heuristic Design
标题：超越节点：自动启发式设计中高效MCTS的分支级选择
链接：https://arxiv.org/abs/2602.00549

作者：Kezhao Lai,Yutao Lai,Hai-Lin Liu

【83】Physiology as Language: Translating Respiration to Sleep EEG
标题：作为语言的生理学：将呼吸转化为睡眠脑电
链接：https://arxiv.org/abs/2602.00526

作者：Kaiwen Zha,Chao Li,Hao He,Peng Cao,Tianhong Li,Ali Mirzazadeh,Ellen Zhang,Jong Woo Lee,Yoon Kim,Dina Katabi
备注：Tech report

【84】Stabilizing Decentralized Federated Fine-Tuning via Topology-Aware Alternating LoRA
标题：通过具有布局意识的交替LoRA稳定分散联邦微调
链接：https://arxiv.org/abs/2602.00451

作者：Xiaoyu Wang,Xiaotian Li,Zhixiang Zhou,Chen Li,Yong Liu
备注：17 Pages

【85】Federated-inspired Single-cell Batch Integration in Latent Space
标题：潜在空间中受联邦启发的单单元批集成
链接：https://arxiv.org/abs/2602.00423

作者：Quang-Huy Nguyen,Zongliang Yue,Hao Chen,Wei-Shinn Ku,Jiaqi Wang

【86】Robustness of AutoML on Dirty Categorical Data
标题：AutoML对肮脏分类数据的鲁棒性
链接：https://arxiv.org/abs/2602.00412

作者：Marcos L. P. Bueno,Joaquin Vanschoren

【87】Variational Approach for Job Shop Scheduling
标题：车间调度的变分方法
链接：https://arxiv.org/abs/2602.00408

作者：Seung Heon Oh,Jiwon Baek,Ki Young Cho,Hee Chang Yoon,Jong Hun Woo

【88】Generalized Inverses of Matrix Products: From Fundamental Subspaces to Randomized Decompositions
标题：矩阵积的广义逆：从基本子空间到随机分解
链接：https://arxiv.org/abs/2602.00386

作者：Michał P. Karpowicz,Gilbert Strang

【89】MATRIX: A Multimodal Benchmark and Post-Training Framework for Materials Science
标题：MATRIX：材料科学的多模式基准和后训练框架
链接：https://arxiv.org/abs/2602.00376

作者：Delia McGrath,Curtis Chong,Rohil Kulkarni,Gerbrand Ceder,Adeesh Kolluru
备注：17 pages, 9 Figures, submitted

【90】Quantum Generator Kernels
标题：量子发生器核
链接：https://arxiv.org/abs/2602.00361

作者：Philipp Altmann,Maximilian Mansky,Maximilian Zorn,Jonas Stein,Claudia Linnhoff-Popien
备注：28 pages, 4 figures, 8 tables, under review

【91】In-Run Data Shapley for Adam Optimizer
标题：Adam Optimizer的运行中数据Shapley
链接：https://arxiv.org/abs/2602.00329

作者：Meng Ding,Zeqing Zhang,Di Wang,Lijie Hu
备注：16 pages

【92】Neural Ising Machines via Unrolling and Zeroth-Order Training
标题：通过展开和零阶训练的神经伊辛机
链接：https://arxiv.org/abs/2602.00302

作者：Sam Reifenstein,Timothee Leleu

【93】Self-Attention at Constant Cost per Token via Symmetry-Aware Taylor Approximation
标题：通过对称性意识泰勒逼近以每个代币的固定成本实现自我注意力
链接：https://arxiv.org/abs/2602.00294

作者：Franz A. Heinsen,Leo Kozachkov
备注：For source code and replication instructions, see https://github.com/glassroom/sata_attention. 12 pages, 6 figures (main); 4 pages, 2 figures (appendix)

【94】DIVERGE: Diversity-Enhanced RAG for Open-Ended Information Seeking
标题：DIVERGE：用于开放式信息搜索的多样性增强RAG
链接：https://arxiv.org/abs/2602.00238

作者：Tianyi Hu,Niket Tandon,Akhil Arora

【95】GRIP2: A Robust and Powerful Deep Knockoff Method for Feature Selection
标题：GRIP 2：一种稳健且强大的特征选择深度模仿方法
链接：https://arxiv.org/abs/2602.00218

作者：Bob Junyi Zou,Lu Tian

【96】Reducing Class-Wise Performance Disparity via Margin Regularization
标题：通过边际正则化减少类间性能差异
链接：https://arxiv.org/abs/2602.00205

作者：Beier Zhu,Kesen Zhao,Jiequan Cui,Qianru Sun,Yuan Zhou,Xun Yang,Hanwang Zhang
备注：To appear in ICLR 2026

【97】MiniTensor: A Lightweight, High-Performance Tensor Operations Library
标题：MiniTensor：一个轻量级、高性能的张量运算库
链接：https://arxiv.org/abs/2602.00125

作者：Soumyadip Sarkar

【98】Why LoRA Resists Label Noise: A Theoretical Framework for Noise-Robust Parameter-Efficient Fine-Tuning
标题：为什么LoRA能抵抗标签噪音：噪音稳健、参数高效微调的理论框架
链接：https://arxiv.org/abs/2602.00084

作者：Brady Steele
备注：14 pages, 7 figures, 7 tables

【99】Lossless Embedding Compression via Spherical Coordinates
标题：通过球坐标实现无损嵌入压缩
链接：https://arxiv.org/abs/2602.00079

作者：Han Xiao

【100】A longitudinal geospatial multimodal dataset of post-discharge frailty, physiology, mobility, and neighborhoods
标题：出院后虚弱、生理、流动性和社区的纵向地理空间多模式数据集
链接：https://arxiv.org/abs/2602.00060

作者：Ali Abedi,Charlene H. Chu,Shehroz S. Khan

【101】RAPTOR-AI for Disaster OODA Loop: Hierarchical Multimodal RAG with Experience-Driven Agentic Decision-Making
标题：RAPTOR-AI用于灾难OODA循环：具有经验驱动的统计决策的分层多模式RAG
链接：https://arxiv.org/abs/2602.00030

作者：Takato Yasuno
备注：4 figures, 3 tables

【102】Provably Data-driven Multiple Hyper-parameter Tuning with Structured Loss Function
标题：具有结构化损失函数的可证明数据驱动的多超参数调整
链接：https://arxiv.org/abs/2602.02406

作者：Tung Quoc Le,Anh Tuan Nguyen,Viet Anh Nguyen

【103】Well-Posed KL-Regularized Control via Wasserstein and Kalman-Wasserstein KL Divergences
标题：通过Wasserstein和Kalman-Wasserstein KL分歧实现良好的KL正规化控制
链接：https://arxiv.org/abs/2602.02250

作者：Viktor Stein,Adwait Datar,Nihat Ay
备注：37 pages, 9 figures, comments welcome

【104】PCA of probability measures: Sparse and Dense sampling regimes
标题：概率测量的PCA：稀疏和密集抽样制度
链接：https://arxiv.org/abs/2602.02190

作者：Gachon Erell,Jérémie Bigot,Elsa Cazelles

【105】Training-free score-based diffusion for parameter-dependent stochastic dynamical systems
标题：参数相关随机动力系统的免训练基于分数的扩散
链接：https://arxiv.org/abs/2602.02113

作者：Minglei Yang,Sicheng He

【106】Stochastic Interpolants in Hilbert Spaces
标题：Hilbert空间中的随机插值
链接：https://arxiv.org/abs/2602.01988

作者：James Boran Yu,RuiKang OuYang,Julien Horwood,José Miguel Hernández-Lobato
备注：8 pages, 1 figure, 2 tables

【107】Privacy Amplification by Missing Data
标题：缺失数据导致隐私扩大
链接：https://arxiv.org/abs/2602.01928

作者：Simon Roburin,Rafaël Pinot,Erwan Scornet

【108】Physics-Informed Chebyshev Polynomial Neural Operator for Parametric Partial Differential Equations
标题：参数偏微方程的物理信息Chebyshev多元神经运算器
链接：https://arxiv.org/abs/2602.01737

作者：Biao Chen,Jing Wang,Hairun Xie,Qineng Wang,Shuai Zhang,Yifan Xia,Jifa Zhang
备注：28pages

【109】Rethinking Multinomial Logistic Mixture of Experts with Sigmoid Gating Function
标题：重新思考具有Sigmoid门控功能的专家的多项逻辑混合
链接：https://arxiv.org/abs/2602.01466

作者：Tuan Minh Pham,Thinh Cao,Viet Nguyen,Huy Nguyen,Nhat Ho,Alessandro Rinaldo
备注：Tuan Minh Pham, Thinh Cao, and Viet Nguyen contributed equally to this work

【110】Online Social Welfare Function-based Resource Allocation
标题：基于网络社会福利功能的资源配置
链接：https://arxiv.org/abs/2602.01400

作者：Kanad Pardeshi,Samsara Foubert,Aarti Singh

【111】SSNAPS: Audio-Visual Separation of Speech and Background Noise with Diffusion Inverse Sampling
标题：SSNAPS：使用扩散反采样的语音和背景噪音的视听分离
链接：https://arxiv.org/abs/2602.01394

作者：Yochai Yemini,Yoav Ellinson,Rami Ben-Ari,Sharon Gannot,Ethan Fetaya

【112】AI Meets Plasticity: A Comprehensive Survey
标题：人工智能遇上可塑性：全面调查
链接：https://arxiv.org/abs/2602.01215

作者：Hadi Bakhshan,Sima Farshbaf,Junior Ramirez Machado,Fernando Rastellini Canela,Josep Maria Carbonell

【113】Sublinear Time Quantum Algorithm for Attention Approximation
标题：注意力逼近的次线性时间量子算法
链接：https://arxiv.org/abs/2602.00874

作者：Zhao Song,Jianfei Xue,Jiahao Zhang,Lichen Zhang
备注：ICLR 2026

【114】Multivariate Time Series Data Imputation via Distributionally Robust Regularization
标题：通过分布稳健正规化进行多元时间序列数据插补
链接：https://arxiv.org/abs/2602.00844

作者：Che-Yi Liao,Zheng Dong,Gian-Gabriel Garcia,Kamran Paynabar

【115】Score-based Metropolis-Hastings for Fractional Langevin Algorithms
标题：基于分数的Metropolis-Hastings分数Langevin算法
链接：https://arxiv.org/abs/2602.00835

作者：Ahmed Aloui,Junyi Liao,Ali Hasan,Jose Blanchet,Vahid Tarokh

【116】Harmful Overfitting in Sobolev Spaces
标题：Sobolev空间中的有害过拟合
链接：https://arxiv.org/abs/2602.00825

作者：Kedar Karhadkar,Alexander Sietsema,Deanna Needell,Guido Montufar

【117】Safety-Efficacy Trade Off: Robustness against Data-Poisoning
标题：安全性与功效权衡：对抗数据中毒的稳健性
链接：https://arxiv.org/abs/2602.00822

作者：Diego Granziol

【118】Zero-Flow Encoders
标题：零流编码器
链接：https://arxiv.org/abs/2602.00797

作者：Yakun Wang,Leyang Wang,Song Liu,Taiji Suzuki
备注：Yakun Wang and Leyang Wang contributed equally to this work

【119】Sampling from multi-modal distributions on Riemannian manifolds with training-free stochastic interpolants
标题：具有免训练随机插值的Riemann上多峰分布的采样
链接：https://arxiv.org/abs/2602.00641

作者：Alain Durmus,Maxence Noble,Thibaut Pellerin

【120】Action-Free Offline-to-Online RL via Discretised State Policies
标题：通过离散化的州政策实现免费的线下到在线RL
链接：https://arxiv.org/abs/2602.00629

作者：Natinael Solomon Neggatu,Jeremie Houssineau,Giovanni Montana
备注：ICLR 2026

【121】Stabilizing Fixed-Point Iteration for Markov Chain Poisson Equations
标题：Markov链Poisson方程的稳定定点迭代
链接：https://arxiv.org/abs/2602.00474

作者：Yang Xu,Vaneet Aggarwal

【122】Exact Instance Compression for Convex Empirical Risk Minimization via Color Refinement
标题：通过颜色细化实现凸经验风险最小化的精确实例压缩
链接：https://arxiv.org/abs/2602.00437

作者：Bryan Zhu,Ziang Chen

【123】Topological Residual Asymmetry for Bivariate Causal Direction
标题：二元因果方向的剩余不对称性
链接：https://arxiv.org/abs/2602.00427

作者：Mouad El Bouchattaoui

【124】Shuffle and Joint Differential Privacy for Generalized Linear Contextual Bandits
标题：广义线性上下文盗贼的洗牌和联合差异隐私
链接：https://arxiv.org/abs/2602.00417

作者：Sahasrajit Sarmasarkar

【125】The GT-Score: A Robust Objective Function for Reducing Overfitting in Data-Driven Trading Strategies
标题：GT得分：一个稳健的目标函数，用于减少数据驱动交易策略中的过度匹配
链接：https://arxiv.org/abs/2602.00080

作者：Alexander Sheppert

【126】On finite-dimensional encoding/decoding theorems for neural operators
标题：神经运算符的有限维编码/解码定理
链接：https://arxiv.org/abs/2602.00068

作者：Vinícius Luz Oliveira,Vladimir G. Pestov
备注：18 pp., 1 figure, latex

机器翻译由腾讯交互翻译提供，仅供参考

点击“阅读原文”获取带摘要的学术速递