cs.LG(2026-06-10)

📊 共 33 篇论文 | 🔗 4 篇有代码

🎯 兴趣领域导航

支柱二:RL算法与架构 (RL & Architecture) (16) 支柱九:具身大模型 (Embodied Foundation Models) (11 🔗3) 支柱一:机器人控制 (Robot Control) (3 🔗1) 支柱八:物理动画 (Physics-based Animation) (2) 支柱五:交互与反应 (Interaction & Reaction) (1)

🔬 支柱二:RL算法与架构 (RL & Architecture) (16 篇)

#题目一句话要点标签🔗
1 Physics-Distilled Neural Network enabled by Large Language Models for Manufacturing Process-Property Predictive Modeling 提出基于大型语言模型的物理蒸馏神经网络以解决制造过程属性预测问题 predictive model distillation large language model
2 RePAIR: Predictive Self-Supervised Representation Learning in Chess 提出RePAIR以解决棋类游戏表示学习问题 reinforcement learning JEPA Joint-Embedding Predictive Architecture
3 IAPO: Input Attribution-Aware Policy Optimization for Tool Use in Small Multimodal Agents 提出IAPO以解决小型多模态代理工具调用能力不足问题 reinforcement learning reward design multimodal
4 Latent World Recovery for Multimodal Learning with Missing Modalities 提出潜在世界恢复框架以解决多模态学习中的缺失模态问题 representation learning multimodal
5 Space-sampled Value Decay: Forgetting Mechanisms for Non-stationary Deep Reinforcement Learning 提出空间采样值衰减机制以解决非平稳深度强化学习问题 reinforcement learning deep reinforcement learning SAC
6 Multimodal Ordinal Modeling of Alzheimer's Disease Severity Using Structural MRI and Clinical Data 提出多模态序数建模方法以自动评估阿尔茨海默病严重性 MAE multimodal
7 RLCSD: Reinforcement Learning with Contrastive On-Policy Self-Distillation 提出RLCSD以解决OPSD中的特权诱导风格漂移问题 reinforcement learning teacher-student distillation
8 APEX: A Network-Native Time-Series Foundation Model for Forecasting and Anomaly Detection for Wireless Edge Operations 提出APEX以解决无线网络遥测预测与异常检测问题 MAE foundation model
9 On Subquadratic Architectures: From Applications to Principles 提出xLSTM以解决变换器模型的计算复杂性问题 Mamba distillation large language model
10 Generalization Hacking: Models Can Game Reinforcement Learning by Preventing Behavioral Generalization 提出一般化黑客技术以解决强化学习中的行为泛化问题 reinforcement learning chain-of-thought
11 APPO: Agentic Procedural Policy Optimization 提出APPO以优化代理强化学习中的决策分支与信用分配问题 reinforcement learning large language model
12 Breaking Entropy Bounds: Accelerating RL Training via MTP with Rejection Sampling 提出Bebop以解决强化学习训练中的MTP接受率瓶颈问题 reinforcement learning large language model
13 Reinforcement Learning Disrupts Gradient-Based Adversarial Optimization 提出强化学习以干扰基于梯度的对抗优化 reinforcement learning
14 Phase Transitions in Attention: A Bayesian Theory of Copy Head Emergence 提出贝叶斯理论解释注意力机制中的复制子电路出现 linear attention large language model
15 Spectrally Regularized Latent Flow Matching for Turbulence Generation 提出光谱正则化潜在流匹配以解决湍流生成问题 flow matching
16 When Context Returns: Toward Robust Internalization in On-Policy Distillation 提出轻量级一致性正则化以解决上下文引起的性能下降问题 distillation

🔬 支柱九:具身大模型 (Embodied Foundation Models) (11 篇)

#题目一句话要点标签🔗
17 Information-Theoretic Decomposition for Multimodal Interaction Learning 提出DMIL以解决多模态学习中的动态交互问题 multimodal
18 Tabular Foundation Models for Clinical Survival Analysis via Survival-Aware Adaptation 提出轻量级适应方法以解决临床生存分析问题 foundation model
19 Time-Series Foundation Model Embeddings for Remaining Useful Life Estimation 提出轻量级学习方法以提升剩余使用寿命预测精度 foundation model
20 Sparse probes and murky physics: a case study of interpretability challenges in a foundation model for continuum dynamics 通过稀疏自编码器探讨基础模型在连续动力学中的可解释性挑战 foundation model
21 TimeRouter: Efficient and Adaptive Routing of Time-Series Foundation Models 提出TimeRouter以解决时间序列模型专家选择问题 foundation model
22 Fine-tuning Multi-modal LLMs with ART: Art-based Reinforcement Training 提出ART方法以优化多模态大语言模型的微调 large language model multimodal
23 Harness In-Context Operator Learning with Chain of Operators 提出链式算子框架以解决ICON泛化不足问题 large language model
24 Bootstrapped Monitoring: Leveraging Transparent Reasoning to Oversee Stronger AI Agents 提出引导监控以解决AI代理监控可靠性问题 chain-of-thought
25 Categorical Prior Lock-in: Why In-Context Learning Fails for Structured Data 提出分类先验锁定以解决结构化数据的学习问题 large language model
26 Beyond representational alignment with brain-guided language models for robust reasoning 提出脑导向语言模型以增强推理能力 large language model
27 Bergson: An Open Source Library for Data Attribution Bergson库的提出以解决数据归因工具缺乏的问题 large language model

🔬 支柱一:机器人控制 (Robot Control) (3 篇)

#题目一句话要点标签🔗
28 PAWS: Preference Learning with Advantage-Weighted Segments 提出PAWS以解决偏好学习中的分布不一致问题 locomotion manipulation reinforcement learning
29 Fourier Features Let Agents Learn High Precision Policies with Imitation Learning 提出傅里叶特征以解决高精度机器人操作问题 manipulation imitation learning
30 Attention by Synchronization in Coupled Oscillator Networks 提出基于耦合振荡器网络的注意力机制以解决能量受限问题 OSC

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

#题目一句话要点标签🔗
31 PCA-Enhanced Adaptive NVAR Framework for High-Resolution Sea Surface Temperature Forecasting in the East Sea 提出基于PCA增强的自适应NVAR框架以提高东海海表温度预测精度 spatiotemporal
32 Urban Heat MiniCubes: An AI-Ready dataset for urban heat research 提出Urban Heat MiniCubes以解决城市热岛效应研究数据不足问题 spatiotemporal

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

#题目一句话要点标签🔗
33 Neuro-Relational Programs: Unifying Queries and Neural Computation over Structured Data 提出神经关系程序以统一结构化数据的查询与神经计算 OMOMO

⬅️ 返回 cs.LG 首页 · 🏠 返回主页