cs.LG（2025-06-14）

📊 共 18 篇论文 | 🔗 1 篇有代码

🎯 兴趣领域导航

支柱二：RL算法与架构 (RL & Architecture) (10) 支柱九：具身大模型 (Embodied Foundation Models) (7 🔗1) 支柱八：物理动画 (Physics-based Animation) (1)

🔬 支柱二：RL算法与架构 (RL & Architecture) (10 篇)

#题目一句话要点标签🔗⭐
1 Generalizable Trajectory Prediction via Inverse Reinforcement Learning with Mamba-Graph Architecture 提出基于Mamba-Graph架构的逆强化学习方法，提升轨迹预测的泛化能力 reinforcement learning inverse reinforcement learning Mamba
2 DR-SAC: Distributionally Robust Soft Actor-Critic for Reinforcement Learning under Uncertainty 提出DR-SAC算法，增强SAC在不确定环境下强化学习的鲁棒性 reinforcement learning deep reinforcement learning SAC
3 Similarity as Reward Alignment: Robust and Versatile Preference-based Reinforcement Learning 提出SARA：一种基于相似性的奖励对齐方法，提升偏好强化学习的鲁棒性和通用性 reinforcement learning offline RL reward shaping
4 Merlin: Multi-View Representation Learning for Robust Multivariate Time Series Forecasting with Unfixed Missing Rates 提出Merlin，通过多视角表征学习增强MTSF模型在非固定缺失率下的鲁棒性。 representation learning contrastive learning distillation
5 Relative Entropy Regularized Reinforcement Learning for Efficient Encrypted Policy Synthesis 提出基于相对熵正则化强化学习的加密策略合成方法，实现高效隐私保护。 reinforcement learning OMOMO
6 PLD: A Choice-Theoretic List-Wise Knowledge Distillation 提出基于选择理论的列表式知识蒸馏方法PLD，提升模型压缩性能 teacher-student distillation
7 Mapping Neural Signals to Agent Performance, A Step Towards Reinforcement Learning from Neural Feedback 提出NEURO-LOOP框架，探索脑信号到智能体性能的映射，为神经反馈强化学习奠定基础 reinforcement learning
8 Information fusion strategy integrating pre-trained language model and contrastive learning for materials knowledge mining 提出融合预训练语言模型与对比学习的信息融合策略，用于材料知识挖掘。 contrastive learning
9 Interpretable Causal Representation Learning for Biological Data in the Pathway Space 提出SENA-discrepancy-VAE，用于生物数据因果表征学习，提升模型可解释性。 representation learning
10 PROTOCOL: Partial Optimal Transport-enhanced Contrastive Learning for Imbalanced Multi-view Clustering 提出PROTOCOL框架，解决不平衡多视图聚类中的表征退化问题。 contrastive learning

🔬 支柱九：具身大模型 (Embodied Foundation Models) (7 篇)

#题目一句话要点标签🔗⭐
11 Unveiling Confirmation Bias in Chain-of-Thought Reasoning 揭示思维链推理中大语言模型的确认偏差现象 large language model chain-of-thought ✅
12 Exploring the Secondary Risks of Large Language Models 探索大语言模型在良性交互下的次生风险，提出SecLens评估框架。 large language model
13 HYPER: A Foundation Model for Inductive Link Prediction with Knowledge Hypergraphs 提出HYPER模型以解决知识超图的归纳链接预测问题 foundation model
14 Beyond Frequency: The Role of Redundancy in Large Language Model Memorization 揭示冗余在大型语言模型记忆中的作用，提出基于冗余的数据预处理方法。 large language model
15 A Framework for Generating Conversational Recommendation Datasets from Behavioral Interactions ConvRecStudio：基于行为交互生成对话式推荐数据集的框架 large language model
16 Automatic Expert Discovery in LLM Upcycling via Sparse Interpolated Mixture-of-Experts 提出SIMoE，通过稀疏插值混合专家模型实现LLM的自动专家发现与能力提升。 large language model
17 QiMeng-Attention: SOTA Attention Operator is generated by SOTA Attention Algorithm 提出QiMeng-Attention，通过LLM自动生成高性能Attention算子，解决长文本场景下的性能瓶颈。 large language model

🔬 支柱八：物理动画 (Physics-based Animation) (1 篇)

#题目一句话要点标签🔗⭐
18 Path-specific effects for pulse-oximetry guided decisions in critical care 利用路径特定效应，研究脉搏血氧仪偏差对重症监护决策的影响 PULSE

⬅️ 返回 cs.LG 首页 · 🏠 返回主页