cs.LG（2026-09-03）

📊 共 18 篇论文 | 🔗 3 篇有代码

🎯 兴趣领域导航

支柱二：RL算法与架构 (RL & Architecture) (11 🔗3) 支柱九：具身大模型 (Embodied Foundation Models) (5) 支柱一：机器人控制 (Robot Control) (2)

🔬 支柱二：RL算法与架构 (RL & Architecture) (11 篇)

#题目一句话要点标签🔗⭐
1 Multi-step Proximal Policy Improvement in Offline Reinforcement Learning 提出多步近端策略改进以解决离线强化学习中的策略更新问题 reinforcement learning TD3 offline reinforcement learning
2 Gradients Know What Outcomes Don't: Unlocking Reinforcement Learning for LLM Reasoning with Gradient-Aligned Rewards 提出梯度对齐奖励以解决强化学习在大语言模型推理中的局限性 reinforcement learning large language model chain-of-thought ✅
3 DE-Venus: A Data-Efficient RLVR Framework for Large Language Models 提出DE-Venus框架以解决RLVR中的数据效率问题 reinforcement learning large language model
4 Out-of-Distribution Generalisation with Sequence Models in Offline Multi-Agent Reinforcement Learning 提出多任务序列模型以解决离线多智能体强化学习中的零-shot任务泛化问题 reinforcement learning zero-shot transfer
5 Risk and Anomaly Identification for Distribution Network Optimal Operation Based on Reinforcement Learning and Uncertainty Quantification 提出基于强化学习和不确定性量化的配电网络风险与异常识别方法 reinforcement learning deep reinforcement learning DRL
6 Subspace Inference Enables Efficient Active Reward Learning from Preferences 提出PreferenceEKF以解决人类偏好学习中的不确定性问题 reinforcement learning offline reinforcement learning preference learning ✅
7 TraveL: Transformer-based Multi-view Path Distributional Representation Learning 提出TraveL框架以解决路径表示学习中的旅行者行为多样性问题 representation learning MAE
8 Unlocking Lossless Speedups in LLMs via Discrete Diffusion 提出扩散增强LLM以解决自回归生成速度慢的问题 distillation large language model ✅
9 A Peer-Relative Representation Learning Framework for Energy Inefficiency Identification in Mobile Network Sites 提出一种同行相对表示学习框架以识别移动网络站点的能效问题 representation learning
10 Beyond Straightness: Non-Crossing Flow Matching via Quantile AlignTree Coupling 提出量化对齐树流匹配以解决流匹配中的路径交叉问题 flow matching
11 EF1-Constrained Nash Social Welfare with Identical Additive Valuations: Complexity, Guarantees, and Experiments 提出PriorityNet以解决EF1约束下的Nash社会福利问题 reinforcement learning deep reinforcement learning

🔬 支柱九：具身大模型 (Embodied Foundation Models) (5 篇)

#题目一句话要点标签🔗⭐
12 Landmark-Based Discrimination of Injury-Associated Athlete-Sessions from Minute-Resolution Multimodal Football Monitoring Data 提出基于地标的模型以解决运动员受伤会话标记问题 multimodal
13 Language-encoded network topology enables large language models to reason about complex networks 提出BioGlyph以解决大语言模型对复杂网络推理的挑战 large language model
14 Headroom-Drift Replay: A Primitive for Principled Replay Control in GRPO 提出Headroom-Drift Replay以解决GRPO中的重放控制问题 multimodal
15 High-Dimensional Learning Dynamics of Attention-Indexed Models 提出高维学习动态模型以解析注意力机制训练过程 foundation model
16 A Large Open Multi-Energy Corpus of Soil Compaction Tests, with Machine-Learning Baselines 提出一个大型开放的土壤压实测试数据集以解决数据稀缺问题 foundation model

🔬 支柱一：机器人控制 (Robot Control) (2 篇)

#题目一句话要点标签🔗⭐
17 Latent Energy Action Planning with World Models 提出LEAP以解决高维观察下的动作规划问题 model predictive control world model worldmodel
18 Spruce: Scalable Private Outsourced Retrieval Using Compact Embeddings 提出Spruce以解决安全外包检索中的效率问题 MPC

⬅️ 返回 cs.LG 首页 · 🏠 返回主页