cs.LG（2026-07-29）

📊 共 28 篇论文 | 🔗 3 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (13 🔗1) 支柱二：RL算法与架构 (RL & Architecture) (11 🔗1) 支柱八：物理动画 (Physics-based Animation) (2 🔗1) 支柱一：机器人控制 (Robot Control) (1) 支柱三：空间感知与语义 (Perception & Semantics) (1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (13 篇)

#题目一句话要点标签🔗⭐
1 Foundation Models for Face Presentation Attack Detection: A Unified Linear-Probing Benchmark 提出统一线性探测基准以提升人脸呈现攻击检测能力 foundation model
2 Using large language models to probe the limits of atom-centered structural descriptors 利用大型语言模型探讨原子中心结构描述符的极限 large language model
3 Harnessing Large Language Models for Intelligent Resource Allocation in the Internet of Everything 提出基于大语言模型的智能资源调度机制以应对物联网挑战 large language model
4 AgentGFM: A Graph Foundation Model with Node-Agent Information-Flow Control 提出AgentGFM以解决图模型信息传播控制问题 foundation model
5 InferScale: GPU-Native KV Injection for Personalized LLM Serving 提出InferScale以解决个性化LLM服务中的延迟问题 large language model
6 Equilibrium Training of Energy-Based Models with Parallel Trajectory Tempering 提出平行轨迹温度调节算法以提升能量模型训练效率 multimodal
7 Thinking Under Uncertainty: Evidence Use and Information-Seeking in Language Models 提出思维机制以提升语言模型在不确定性下的决策能力 large language model
8 Budget-Aware LLM Discovery via Cost-Calibrated Frontier Utility 提出CostAda以解决预算感知的LLM发现问题 large language model
9 Forecasting Trajectory-Level Safety Risks in Black-Box Multi-Turn Interactions 提出Recast框架以解决LLM安全风险预测问题 large language model
10 AIGen: Automating AI Bill of Materials Generation Through Hybrid MLOps Integration 提出AIGen以自动生成符合标准的AI材料清单 large language model ✅
11 From Tokens to Watt-hours: Analytical Energy Estimation for LLM Inference on Modern GPUs 提出一种分析性能量估算方法以优化LLM推理能耗 large language model
12 Neural Architecture Search for Traffic Prediction: A Survey of Methods, Challenges, and Future Directions 通过神经架构搜索优化交通预测模型设计 foundation model
13 ClockRoPE: Random Fourier Rotations for Temporal Routine Modeling 提出ClockRoPE以解决序列推荐中的时间周期建模问题 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🔗⭐
14 What Can Latent World Models Know? Physical Parameter Identifiability in Multimodal Predictive Representations 提出证书门控协议以识别潜在世界模型中的物理参数 world model world models multimodal
15 Temporally Centered SIGReg Improves Multi-Task LeWorldModel Learning: From Analysis to Method 提出时序中心化SIGReg以改善多任务LeWorldModel学习 behavior cloning diffusion policy world model
16 HiFloat4 Format for End-To-End Reinforcement Learning Post-Training of Large Language Models 提出HiFloat4格式与Rollout Residual Quantization以解决FP4强化学习后训练精度问题 reinforcement learning large language model
17 CalTwin: Towards Calibrated, Shift-Robust Medical World Models via Fisher-Information Regularisation 提出CalTwin以解决医疗世界模型的校准与鲁棒性问题 world model world models latent dynamics
18 DHRCL:Training Code LLMs with Dense Hierarchical Rewards and Curriculum Learning 提出DHRCL框架以优化代码大语言模型的训练效果 reinforcement learning curriculum learning large language model
19 CoCaRS: Correlation Calibration-Based Redundancy Suppression for Heterogeneous Knowledge Distillation 提出CoCaRS以解决异构知识蒸馏中的冗余抑制问题 SAC teacher-student distillation
20 SkillRise: Agentic Reinforcement Learning for Cross-Task Skill Evolution 提出SkillRise以解决跨任务技能演化问题 reinforcement learning large language model
21 Collaborative Weighting with Pessimistic Critic for Mitigating Overestimation in Off-Policy Reinforcement Learning 提出CWAC以解决离线强化学习中的过度估计问题 reinforcement learning SAC TD3
22 DREvo: Distilling Recalibrated Historical Experience for Harness Self-Evolution 提出DREvo以解决历史经验在自我进化中的有效性问题 distillation large language model
23 Learning Dynamic User Personas from Implicit Interaction Streams via Iterative Refinement 提出IRIS框架以解决个性化语言模型的用户建模问题 preference learning large language model
24 Single-Beat Cuffless Blood Pressure Estimation Using Ear-PPG and ECG with a Lightweight Hybrid Learning Framework 提出轻量级混合学习框架以实现无袖带血压监测 MAE PULSE ✅

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

#题目一句话要点标签🔗⭐
25 PowerAtlas: Towards Electricity-Computing Co-Scheduling for Power Systems 提出PowerAtlas以解决电力与计算调度协同问题 spatiotemporal large language model ✅
26 Skillful forecasting of offshore winds from satellite scatterometer constellations 提出WindCastNet以解决海上风速短期预测问题 spatiotemporal

🔬 支柱一：机器人控制 (Robot Control) (1 篇)

#题目一句话要点标签🔗⭐
27 SCOUT: Per-Context Reset Curricula for Sparse-Reward Reinforcement Learning 提出SCOUT以解决稀疏奖励强化学习中的重置课程问题 manipulation reinforcement learning

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

#题目一句话要点标签🔗⭐
28 ToxScreen: Detecting Whether an LLM Has Been Poisoned 提出ToxScreen以检测大语言模型是否被毒化 affordance large language model

⬅️ 返回 cs.LG 首页 · 🏠 返回主页