cs.LG（2025-06-07）

📊 共 9 篇论文 | 🔗 2 篇有代码

🎯 兴趣领域导航

支柱二：RL算法与架构 (RL & Architecture) (4 🔗1) 支柱九：具身大模型 (Embodied Foundation Models) (4 🔗1) 支柱五：交互与反应 (Interaction & Reaction) (1)

🔬 支柱二：RL算法与架构 (RL & Architecture) (4 篇)

#	题目	一句话要点	标签	🔗	⭐
1	Is Optimal Transport Necessary for Inverse Reinforcement Learning?	提出两种简单启发式奖励函数，挑战逆强化学习中优化传输的必要性	reinforcement learning inverse reinforcement learning
2	Curriculum Reinforcement Learning from Easy to Hard Tasks Improves LLM Reasoning	提出E2H Reasoner，通过课程强化学习提升LLM的推理能力	reinforcement learning curriculum learning	✅
3	Can In-Context Reinforcement Learning Recover From Reward Poisoning Attacks?	提出对抗训练决策预训练Transformer，提升ICRL在奖励中毒攻击下的鲁棒性	reinforcement learning
4	Learning Robust Heterogeneous Graph Representations via Contrastive-Reconstruction under Sparse Semantics	HetCRF：通过对比重构学习异构图的鲁棒表示，解决语义稀疏问题	masked autoencoder MAE contrastive learning

🔬 支柱九：具身大模型 (Embodied Foundation Models) (4 篇)

#	题目	一句话要点	标签	🔗	⭐
5	Breaking Data Silos: Towards Open and Scalable Mobility Foundation Models via Generative Continual Learning	MoveGCL：通过生成式持续学习打破数据孤岛，实现开放且可扩展的出行基础模型	foundation model	✅
6	SAFER: A Calibrated Risk-Aware Multimodal Recommendation Model for Dynamic Treatment Regimes	提出SAFER以解决动态治疗方案中的风险管理问题	multimodal
7	Basis Transformers for Multi-Task Tabular Regression	提出Basis Transformers，解决表格回归中的异构结构和缺失信息问题	large language model
8	MarginSel : Max-Margin Demonstration Selection for LLMs	MarginSel：面向大语言模型的最大间隔示范选择方法	large language model

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

#	题目	一句话要点	标签	🔗	⭐
9	Caterpillar GNN: Replacing Message Passing with Efficient Aggregation	提出Caterpillar GNN，通过高效聚合替代消息传递，提升图神经网络性能。	OMOMO

⬅️ 返回 cs.LG 首页 · 🏠 返回主页