cs.LG(2026-06-18)

📊 共 29 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) (17 🔗1) 支柱九:具身大模型 (Embodied Foundation Models) (8 🔗1) 支柱八:物理动画 (Physics-based Animation) (2 🔗1) 支柱七:动作重定向 (Motion Retargeting) (1) 支柱三:空间感知与语义 (Perception & Semantics) (1)

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

#题目一句话要点标签🔗
1 Manifold Bandits: Bayesian Curriculum Learning over the Latent Geometry of Large Language Models 提出贝叶斯流形课程学习以优化大语言模型的推理能力 reinforcement learning curriculum learning large language model
2 Direct Advantage Estimation for Scalable and Sample-efficient Deep Reinforcement Learning 提出直接优势估计以解决深度强化学习的样本效率问题 reinforcement learning deep reinforcement learning latent dynamics
3 Sensorimotor World Models: Perception for Action via Inverse Dynamics 提出传感运动世界模型以解决表示崩溃和动作对齐问题 world model world models JEPA
4 Uncertainty-Aware Reward Modeling for Stable RLHF 提出不确定性感知奖励建模以解决RLHF中的奖励不可靠问题 reinforcement learning RLHF large language model
5 Multi-Modal Contrastive Learning for Implicit Earth Embeddings via Location Tying 提出多模态对比学习方法以解决地理位置嵌入问题 contrastive learning multimodal
6 Connect the Dots: Training LLMs for Long-Lifecycle Agents with Cross-Domain Generalization Via Reinforcement Learning 提出CoD框架以提升长生命周期智能体的跨域泛化能力 reinforcement learning large language model
7 CRAX: Fast Safe Reinforcement Learning Benchmarking 提出CRAX以解决安全强化学习基准测试的计算效率问题 reinforcement learning curriculum learning
8 A Model-Driven Approach for Developing Families of Reinforcement Learning Environments 提出模型驱动方法以高效开发强化学习环境家族 reinforcement learning curriculum learning
9 VIMPO: Value-Implicit Policy Optimization for LLMs 提出VIMPO以解决大语言模型的奖励分配问题 reinforcement learning PPO large language model
10 SL-S4Wave: Self-Supervised Learning of Physiological Waveforms with Structured State Space Models 提出SL-S4Wave以解决多通道生理波形建模问题 state space model contrastive learning
11 OnDeFog: Online Decision Transformer under Frame Dropping 提出OnDeFog以解决帧丢失下的决策问题 reinforcement learning decision transformer
12 Boundary Embedding Shaping with Adaptive Contrastive Learning for Graph Structural Disentanglement 提出边界嵌入塑形以解决图结构纠缠问题 contrastive learning
13 Quantile of Means: A Bonus-Free Ensemble Method for Minimax Optimal Reinforcement Learning 提出无奖励的集成方法以解决最小最大最优强化学习问题 reinforcement learning
14 Hierarchical Control in Multi-Agent Games: LLM-based Planning and RL Execution 提出层次控制架构以解决多智能体游戏中的协调问题 reinforcement learning large language model
15 StreamKL: Fast and Memory-Efficient KL Divergence for Boosting Attention Distillation 提出StreamKL以解决注意力蒸馏中的内存和速度问题 distillation
16 Multi-Granular Attention-Driven Reinforcement Learning Framework for Web Intelligent Enhancement Systems 提出多粒度注意力驱动的强化学习框架以提升网络智能增强系统 reinforcement learning
17 Off-Policy Evaluation for Missingness-Aware Policies in MDPs with Rewards Missing Not at Random 提出缺失奖励的离线强化学习评估方法以解决选择偏差问题 reinforcement learning offline reinforcement learning

🔬 支柱九:具身大模型 (Embodied Foundation Models) (8 篇)

#题目一句话要点标签🔗
18 What Makes Effective Supervision in Latent Chain-of-Thought: An Information-Theoretic Analysis 提出信息论分析以优化潜在链式思维的监督机制 chain-of-thought
19 Alzheimer's Disease Diagnosis using a Multimodal Approach with 3D MRI and PET 提出多模态方法结合3D MRI与PET以改善阿尔茨海默病诊断 multimodal
20 Efficiently Representing Algorithms With Chain-of-Thought Transformers 提出链式思维变换器以高效模拟Word RAM算法 chain-of-thought
21 Contagion Networks: Evaluator Bias Propagation in Multi-Agent LLM Systems 提出传播网络框架以解决多智能体系统中的评估者偏差问题 large language model
22 Towards Modality-imbalanced Federated Graph Learning: A Data Synthesis-based Approach 提出FedMGS以解决多模态不平衡的联邦图学习问题 multimodal
23 Exploring the potential of AlphaEarth and TESSERA embeddings for Fine-scale Local Climate Zone Mapping: A case study across five cities in Switzerland 利用AlphaEarth和TESSERA嵌入提升细尺度局部气候区映射精度 foundation model
24 Online Dynamic Batching with Formal Guarantees for LLM Training 提出在线动态批处理以解决LLM训练中的效率问题 multimodal
25 ADaPT: Token-Level Decoupling for Efficient Large Reasoning Models 提出ADaPT以解决大规模推理模型的效率与正确性问题 chain-of-thought

🔬 支柱八:物理动画 (Physics-based Animation) (2 篇)

#题目一句话要点标签🔗
26 Multi-Task Bayesian In-Context Learning 提出多任务贝叶斯上下文学习以解决不确定性量化问题 spatiotemporal
27 Enhancing Graph Neural Networks Using Proximity Graphs for Dust Source Emission Forecasting 通过邻近图增强图神经网络以预测尘埃源排放 spatiotemporal

🔬 支柱七:动作重定向 (Motion Retargeting) (1 篇)

#题目一句话要点标签🔗
28 Towards Graph-Based Deep Learning for Map Generalization: Insights from Building Footprints Simplification and Aggregation 提出图神经网络以解决地图泛化中的建筑轮廓简化与聚合问题 spatial relationship

🔬 支柱三:空间感知与语义 (Perception & Semantics) (1 篇)

#题目一句话要点标签🔗
29 How Transparent is DiffusionGemma? 研究DiffusionGemma的透明性以提升模型可解释性 affordance

⬅️ 返回 cs.LG 首页 · 🏠 返回主页