cs.LG（2026-06-30）

📊 共 27 篇论文 | 🔗 5 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (12 🔗4) 支柱二：RL算法与架构 (RL & Architecture) (11 🔗1) 支柱一：机器人控制 (Robot Control) (3) 支柱七：动作重定向 (Motion Retargeting) (1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (12 篇)

#题目一句话要点标签🔗⭐
1 CoMet: Context and Multiplicity Decomposition for Multimodal Uncertainty Estimation 提出CoMet以解决多模态不确定性估计问题 large language model multimodal ✅
2 Relational and Sequential Conformal Inference for Energy Time Series over Graphs via Foundation Models 提出STOIC框架以解决能源时间序列的不确定性预测问题 foundation model
3 An Empirical Study of Security Calibration in Large Language Models for Code 提出大规模实证研究以评估大语言模型的安全校准问题 large language model
4 Bridging the Gap Between Latent and Explicit Reasoning with Looped Transformers 提出LOTUS以解决显性推理与隐性推理之间的差距问题 chain-of-thought
5 RaBitQCache: Rotated Binary Quantization for KVCache in Long Context LLM Inference 提出RaBitQCache以解决长上下文LLM推理中的KV缓存瓶颈问题 large language model ✅
6 TabPATE: Differentially Private Tabular In-Context Learning Without Public Data 提出TabPATE以解决无公共数据的差分隐私表格学习问题 foundation model
7 Calibrating the Evaluator: Does Probability Calibration Mitigate Preference Coupling in LLM Agent Feedback Loops? 提出概率校准以缓解LLM代理反馈循环中的偏好耦合问题 large language model
8 The Calibration Turn in AI-Assisted Research: A Conceptual and Methodological Framework for Evidence-Licensed Claims 提出证据许可声明的框架以提升AI辅助研究的可靠性 foundation model
9 Probing Memorization of Tabular In-Context Learning 提出ICLMEM框架以探测表格模型的记忆化问题 foundation model
10 ComplianceGate: Classifier-Gated Multi-Tier LLM Routing for Inference in Regulated Industries 提出ComplianceGate以解决合规性与成本效率问题 large language model
11 PPT-Eval: A Benchmark for Computer-Use Agents on PowerPoint Tasks 提出PPT-Eval基准以评估计算机使用代理在PowerPoint任务中的表现 multimodal ✅
12 Fora: From Weight-Space to Function-Space Protection in Capability-Preserving Fine-Tuning 提出FORA以解决大语言模型能力保持问题 large language model ✅

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

#题目一句话要点标签🔗⭐
13 Offline Reinforcement Learning for Fluid Controls: Data-based Multi-observational Policy Extraction 提出离线强化学习框架以优化流体控制问题 reinforcement learning deep reinforcement learning offline RL
14 FedLAB: Traceable Semantic Codebooks for Federated Multimodal Graph Foundation Learning 提出FedLAB框架以解决联邦多模态图基础学习中的语义可追溯性问题 representation learning foundation model multimodal
15 Knowledge Distillation from Large Reasoning Models to Compact Student Models: A Case Study on the John O Bryan Mathematics Competition 提出知识蒸馏方法以提升紧凑模型在数学推理中的表现 distillation chain-of-thought
16 TRIAGE: Role-Typed Credit Assignment for Agentic Reinforcement Learning 提出TRIAGE框架以优化代理强化学习中的信用分配问题 reinforcement learning
17 Making Sense of Touch from the Child's View for Contrastive Learning 提出儿童视角的触觉理解方法以促进对比学习 contrastive learning
18 Geometry-Preserving Orthonormal Initialization for Low-Rank Adaptation in RLVR 提出几何保持正交初始化以解决RLVR中的低秩适应问题 reinforcement learning large language model ✅
19 Addressing Over-Refusal in LLMs with Competing Rewards 提出SEAR模型以解决语言模型的过度拒绝问题 reinforcement learning chain-of-thought
20 Diffusing Blame: Task-Dependent Credit Assignment in Biologically Plausible Dual-Stream Networks 提出模态误差路由以解决生物学可行的双流网络中的信用分配问题 reinforcement learning PPO representation learning
21 Improving Certified Robustness via Adversarial Distillation 提出AD-CERT以提升神经网络的认证鲁棒性 distillation
22 Probabilistic Inversion with Flow Matching 提出流匹配方法以解决地球物理中的概率反演问题 flow matching
23 AETDICE: Unified Framework and Offline Optimization for Nonlinear Multi-Objective RL 提出AETDICE以解决非线性多目标强化学习问题 reinforcement learning offline RL

🔬 支柱一：机器人控制 (Robot Control) (3 篇)

#题目一句话要点标签🔗⭐
24 AdaJEPA: An Adaptive Latent World Model 提出AdaJEPA以解决高维观察下的规划失败问题 MPC model predictive control world model
25 Deep Reinforcement Learning for Spacecraft Attitude Control During Atmospheric Re-Entry 提出深度强化学习以解决航天器大气再入姿态控制问题 dynamics randomization reinforcement learning deep reinforcement learning
26 Warp RL: Reshaping Base Policy Distributions for Dynamics Adaptation 提出Warp RL以解决动态适应中的分布形状限制问题 manipulation sim-to-real reinforcement learning

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

#题目一句话要点标签🔗⭐
27 Visualizing High-Dimensional Graph Embeddings via Informed Multi-View Projections 提出高维图嵌入可视化方法以优化2D视角选择 spatial relationship

⬅️ 返回 cs.LG 首页 · 🏠 返回主页