cs.LG（2026-06-04）

📊 共 43 篇论文 | 🔗 4 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (19 🔗2) 支柱二：RL算法与架构 (RL & Architecture) (17 🔗2) 支柱八：物理动画 (Physics-based Animation) (3) 支柱三：空间感知与语义 (Perception & Semantics) (2) 支柱五：交互与反应 (Interaction & Reaction) (1) 支柱一：机器人控制 (Robot Control) (1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (19 篇)

#题目一句话要点标签🔗⭐
1 Bridging Domain Expertise and Generalization for Performance Estimation 提出FRAP以解决分布转移下的性能估计问题 foundation model
2 Subspace-Aware Sparse Autoencoders for Effective Mechanistic Interpretability 提出子空间感知稀疏自编码器以解决特征分裂问题 large language model
3 Learning to Route LLMs from Implicit Cost-Performance Preferences via Meta-Learning 提出MetaRouter以解决个性化LLM路由中的成本与性能优化问题 large language model
4 The Identity Trap in EEG Foundation Models: A Diagnostic Audit 提出FMScope以解决EEG基础模型中的身份陷阱问题 foundation model
5 Generative Criticality in Large Language Model Temperature Scaling 提出统计场框架以分析大语言模型的温度标定问题 large language model
6 TS-ICL: A Flexible Time-Indexed Foundation Model for Time Series via In-Context Learning 提出TS-ICL以解决时间序列建模中的预测与缺失值填补问题 foundation model
7 FAIR-Calib: Frontier-Aware Instability-Reweighted Calibration for Post-Training Quantization of Diffusion Large Language Models 提出FAIR-Calib以解决扩散大语言模型的量化不稳定性问题 large language model
8 Elmes*: Automated Construction of Fine-Grained Evaluation Rubrics for Large Language Models in Long-Tail Educational Scenarios 提出Elmes*框架以解决教育场景中LLM评估的不足问题 large language model
9 CaliDist: Calibrating Large Language Models via Behavioral Robustness to Distraction 提出CaliDist以解决大语言模型的行为鲁棒性问题 large language model
10 The Geography of Algorithmic Judgment: LLM Intermediaries, Place Identity, and Racial Steering in Housing Search 通过行为审计揭示LLM在住房搜索中的种族引导问题 large language model
11 Skip a Layer or Loop It? Learning Program-of-Layers in LLMs 提出动态层程序以提升大语言模型推理效率 large language model
12 The Post-GCN Decade Revisited: Curvature-Stratified Evaluation of Relational Learning 提出曲率分层评估框架以解决关系学习中的偏差问题 foundation model ✅
13 Tangram: Unlocking Non-Uniform KV Cache for Efficient Multi-turn LLM Serving 提出Tangram以解决非均匀KV缓存在多轮LLM服务中的效率问题 large language model ✅
14 GenAutoML: An Agentic Framework for Dynamic Architecture Generation and Optimization in Time-Series Analysis 提出GenAutoML以解决时间序列分析中的动态架构生成问题 large language model
15 Domain-Adapted Small Language Models with Hybrid Post-Processing: Achieving Cost-Efficient, Low-Latency Multi-Label Structured Prediction via LoRA Fine-Tuning on Scarce Data 提出混合后处理的领域适应小语言模型以解决合规评估问题 large language model
16 Automated Proving of Shannon-Type Entropy Inequalities via Fine-Tuned Language Models and Guided Tree Search 提出基于微调语言模型的自动化香农熵不等式证明方法 large language model
17 CASS-RTL: Correctness-Aware Subspace Steering for RTL Generation with LLMs 提出CASS-RTL以解决RTL生成中的正确性问题 large language model
18 Q-GNN: Query-Conditioned Graph Neural Networks with Type Awareness for Knowledge Graph Completion 提出Q-GNN以解决知识图谱补全中的信息利用不足问题 large language model
19 CLaaS: Continual learning as a service for sample efficient online learning 提出CLaaS以解决动态环境中的持续学习问题 large language model

🔬 支柱二：RL算法与架构 (RL & Architecture) (17 篇)

#题目一句话要点标签🔗⭐
20 PAMF: Prior-Aware Multimodal Fusion for Incomplete Time Series Data 提出PAMF以解决多模态时间序列数据缺失问题 flow matching multimodal
21 MDP-GRPO: Stabilized Group Relative Policy Optimization for Multi-Constraint Instruction Following 提出MDP-GRPO以解决多约束指令跟随中的不稳定性问题 reinforcement learning instruction following
22 RREDCoT: Segment-Level Reward Redistribution for Reasoning Models 提出RREDCoT以解决推理模型中的延迟奖励问题 reinforcement learning chain-of-thought
23 OPRD: On-Policy Representation Distillation 提出OPRD以解决现有蒸馏方法的局限性 distillation ✅
24 Representation Learning Enables Scalable Multitask Deep Reinforcement Learning 提出基于表示学习的MR.Q算法以解决多任务深度强化学习的可扩展性问题 reinforcement learning deep reinforcement learning world model
25 Autoregressive Diffusion World Models for Off-Policy Evaluation of LLM Agents 提出自回归扩散世界模型以解决LLM代理的离线评估问题 world model world models large language model
26 Spatiotemporal Imputation with Graph-Informed Flow Matching 提出GiFlow框架以解决时空数据缺失问题 flow matching spatiotemporal ✅
27 HoT-SSM:Higher-order Temporal Knowledge Graph Reasoning with State Space Models for Health Care 提出HoT-SSM以解决医疗知识图谱中的高阶时间推理问题 SSM state space model representation learning
28 Uncertainty-Aware LLM-Guided Policy Shaping for Sparse-Reward Reinforcement Learning 提出不确定性感知的LLM引导策略塑造以解决稀疏奖励问题 reinforcement learning PPO large language model
29 Compress-Distill: Reasoning Trace Compression for Efficient Knowledge Distillation 提出压缩推理轨迹以提高知识蒸馏效率 distillation chain-of-thought
30 Capturing non-Markovian dynamics in non-equilibrium stochastic systems using flow matching 提出流匹配方法以捕捉非马尔可夫动态 flow matching
31 Principles and Practice of Deep Representation Learning: or a Mathematical Theory of Memory 提出深度表示学习原理以解决深度学习模型可解释性问题 representation learning
32 Maximising the Set-Piece Return: Optimising Football Corner Tactics with Graph Reinforcement Learning 提出图结构强化学习优化足球角球战术 reinforcement learning
33 Discrete Causal Representations from Heterogeneous Domains: A Bayesian Approach with Social Survey Applications 提出贝叶斯方法以从异构数据中学习离散因果表示 representation learning multimodal
34 Drag reduction or reward hacking? Recurrent multi-agent reinforcement learning that earns its reward 提出改进的多智能体强化学习以解决奖励偏差问题 reinforcement learning
35 Online KL-Regularized Reinforcement Learning with Function Approximation under Misspecification 提出KL正则化方法以解决模型误设定下的强化学习问题 reinforcement learning
36 Learn to Match: Two-Sided Matching with Temporally Extended Feedback 提出基于时序扩展反馈的双边匹配框架以解决动态匹配问题 reinforcement learning PPO

🔬 支柱八：物理动画 (Physics-based Animation) (3 篇)

#题目一句话要点标签🔗⭐
37 DAS-PINNs for high-dimensional partial differential equations: extending deep adaptive sampling to spacetime domains 提出DAS-PINNs以解决高维时变偏微分方程问题 spatiotemporal
38 Learning to model pediatric asthma exacerbation from multiple risk factors: a case study in coastal Virginia 提出多因素模型以预测儿童哮喘加重事件 spatiotemporal
39 Field Validation of a Multi-Resolution ConvLSTM Framework for Retaining Wall Deformation Prediction 提出多分辨率ConvLSTM框架以预测挡土墙变形 spatiotemporal

🔬 支柱三：空间感知与语义 (Perception & Semantics) (2 篇)

#题目一句话要点标签🔗⭐
40 Knowledge Manifold: A Riemannian Geometric Framework for Semantic Mapping and Geodesic Analysis of Scientific Literature 提出知识流形框架以实现科学文献的语义映射与测地分析 semantic mapping semantic map
41 What Objects Enable, Not What They Are: Functional Latent Spaces for Affordance Reasoning 提出A4D以解决机器人规划中的功能推理问题 affordance

🔬 支柱五：交互与反应 (Interaction & Reaction) (1 篇)

#题目一句话要点标签🔗⭐
42 Design a Reliable LLM-Integrated Interface for Mortality Forecasting 提出可靠的LLM集成界面以改善死亡率预测 OMOMO large language model

🔬 支柱一：机器人控制 (Robot Control) (1 篇)

#题目一句话要点标签🔗⭐
43 Performance Variation in Deep Reinforcement Learning 提出百分位统计方法以解决深度强化学习性能波动问题 MPC reinforcement learning deep reinforcement learning

⬅️ 返回 cs.LG 首页 · 🏠 返回主页