cs.LG（2026-02-25）

📊 共 17 篇论文 | 🔗 6 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (9 🔗3) 支柱二：RL算法与架构 (RL & Architecture) (5 🔗1) 支柱一：机器人控制 (Robot Control) (2 🔗1) 支柱八：物理动画 (Physics-based Animation) (1 🔗1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (9 篇)

#题目一句话要点标签🔗⭐
1 DualWeaver: Synergistic Feature Weaving Surrogates for Multivariate Forecasting with Univariate Time Series Foundation Models DualWeaver：利用协同特征编织代理，增强单变量时间序列基础模型在多元预测中的能力 foundation model ✅
2 TiMi: Empower Time Series Transformers with Multimodal Mixture of Experts 提出TiMi：利用多模态混合专家模型增强时间序列Transformer，提升预测精度。 multimodal
3 Multimodal Survival Modeling and Fairness-Aware Clinical Machine Learning for 5-Year Breast Cancer Risk Prediction 提出一种多模态生存建模框架，用于乳腺癌五年生存风险预测，并关注公平性。 multimodal
4 Extending Sequence Length is Not All You Need: Effective Integration of Multimodal Signals for Gene Expression Prediction Prism框架：有效整合多模态信号，提升基因表达预测精度，无需过度依赖长序列 multimodal
5 Reasoning-Driven Design of Single Atom Catalysts via a Multi-Agent Large Language Model Framework 提出MAESTRO框架以发现高性能单原子催化剂 large language model
6 DHP: Efficient Scaling of MLLM Training with Dynamic Hybrid Parallelism 提出动态混合并行策略以解决多模态大语言模型训练效率问题 large language model multimodal
7 From Words to Amino Acids: Does the Curse of Depth Persist? 揭示蛋白质语言模型深度诅咒：后期层贡献递减，效率待提升 large language model multimodal
8 Muon+: Towards Better Muon via One Additional Normalization Step Muon+：通过额外的归一化步骤提升Muon优化器性能 large language model ✅
9 Learning Recursive Multi-Scale Representations for Irregular Multivariate Time Series Forecasting 提出ReIMTS，通过递归多尺度建模解决不规则多元时间序列预测问题。 TAMP ✅

🔬 支柱二：RL算法与架构 (RL & Architecture) (5 篇)

#题目一句话要点标签🔗⭐
10 Generalisation of RLHF under Reward Shift and Clipped KL Regularisation 针对奖励漂移和KL裁剪正则化的RLHF泛化理论研究 reinforcement learning RLHF large language model
11 GradAlign: Gradient-Aligned Data Selection for LLM Reinforcement Learning 提出GradAlign，通过梯度对齐进行LLM强化学习的数据选择，提升训练稳定性和性能。 reinforcement learning large language model ✅
12 Provable Last-Iterate Convergence for Multi-Objective Safe LLM Alignment via Optimistic Primal-Dual 提出乐观原始-对偶算法，解决多目标安全LLM对齐的末次迭代收敛问题 reinforcement learning RLHF large language model
13 Hierarchical Lead Critic based Multi-Agent Reinforcement Learning 提出基于分层领导评论家的多智能体强化学习方法，提升协作任务性能。 reinforcement learning
14 Mamba Meets Scheduling: Learning to Solve Flexible Job Shop Scheduling with Efficient Sequence Modeling 利用Mamba序列建模高效求解柔性作业车间调度问题 Mamba

🔬 支柱一：机器人控制 (Robot Control) (2 篇)

#题目一句话要点标签🔗⭐
15 Breaking Semantic-Aware Watermarks via LLM-Guided Coherence-Preserving Semantic Injection 提出CSI攻击以破解语义水印的安全性问题 manipulation large language model
16 Learning in the Null Space: Small Singular Values for Continual Learning NESS：利用小奇异值空间进行持续学习，缓解灾难性遗忘。 manipulation ✅

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

#题目一句话要点标签🔗⭐
17 Learning from Yesterday's Error: An Efficient Online Learning Method for Traffic Demand Prediction 提出FORESEE，用于解决交通需求预测中在线适应的计算成本问题。 spatiotemporal foundation model ✅

⬅️ 返回 cs.LG 首页 · 🏠 返回主页