cs.LG（2026-10-09）

📊 共 42 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) (21 🔗1) 支柱二：RL算法与架构 (RL & Architecture) (17 🔗1) 支柱八：物理动画 (Physics-based Animation) (2 🔗1) 支柱一：机器人控制 (Robot Control) (1) 支柱四：生成式动作 (Generative Motion) (1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (21 篇)

#题目一句话要点标签🔗⭐
1 SeoulMMOD: A Large-Scale Multimodal Origin-Destination Flow Benchmark 提出SeoulMMOD以解决城市多模态出行流量预测问题 multimodal
2 On the Encoding of Calendar Features in Covariate-Informed Time Series Foundation Models 提出日历特征编码方法以提升时间序列模型预测精度 foundation model
3 TabReliBench: Evaluating Tabular Foundation Models Beyond Predictive Accuracy 提出TabReliBench以评估表格基础模型的可靠性 foundation model ✅
4 Model-free Bayesian Clustering and Multimodal Posterior Summarisation 提出无模型贝叶斯聚类框架以解决噪声不确定性问题 multimodal
5 MHR-Router: Multimodal Heterogeneous Retrieval-Enhanced Adaptive Routing for VLM Inference 提出MHR-Router以解决VLM推理中的专家选择问题 multimodal
6 Tabular Foundation Models Are Effectively Shallow 提出一种简化的标签无关仿射转换器以优化表格基础模型 foundation model
7 Monarch-MoE: Structured Experts for Multimodal Continual Instruction Tuning 提出Monarch-MoE以解决多模态持续指令调优中的灾难性遗忘问题 multimodal
8 How Much Computation Does Every Forecast Need? Structured Block Skipping in Two Time-Series Foundation Models 提出结构化块跳过方法以优化时间序列预测模型的计算需求 foundation model
9 Scaling RNA Foundation Models through Iterative Refinement 提出Eco-RNA模型以提升RNA功能预测的准确性 foundation model
10 Evidential Multimodal Transformers with Dempster-Shafer Fusion for Emotion Recognition 提出证据多模态变换器以解决情感识别中的不确定性问题 multimodal
11 GIST: Generator-state-Informed Supervision for Tabular Foundation Models 提出GIST以解决表格基础模型的标签不确定性问题 foundation model
12 HART: Hybrid Attention-RWKV Time series Foundation Model 提出HART模型以解决RWKV在长时间序列预测中的性能衰退问题 foundation model
13 QTestBench: A Test Framework for Evaluating LLM-Generated Quantum Code 提出QTestBench框架以评估LLM生成的量子代码测试 large language model
14 Rationally Linearized MLPs 提出理性线性化多层感知机以提高推理效率 large language model
15 Simulating Training Spikes of Deep Neural Networks by Layer-wise Random Feature Models 通过层级随机特征模型模拟深度神经网络训练中的损失波动 large language model
16 PDE-PFN: Prior-Data Fitted Neural PDE Solver 提出PDE-PFN以解决偏微分方程求解问题 large language model
17 Fast Gradient-Based Approximation of Continuous Dynamic Time Warping 提出FastCDTW以提高动态时间规整的计算效率 TAMP
18 Human and LLM Facilitation in Collaborative Design Meetings: A Computational Behavioral Comparison 比较人类与LLM在协作设计会议中的引导行为差异 large language model
19 CB-Orchestrator: Adaptive Workflow Optimization for LLM Agents via Contextual Bandits 提出CB-Orchestrator以解决LLM代理工作流优化问题 large language model
20 Diffusion Precipitation Generation with Medium-range Weather Forecasting 提出动态降水解耦框架以改善天气预报中的降水预测 foundation model
21 Robust Langevin Sampling under State-Dependent Heavy-Tailed Noise with Applications to LLM Fine-Tuning 提出鲁棒Langevin采样以解决状态依赖重尾噪声问题 large language model

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

#题目一句话要点标签🔗⭐
22 EmoGavel: Grounding-Aware Reinforcement Learning with Verifiable Rewards for Multimodal Emotion Preference Learning 提出EmoGavel以解决多模态情感偏好学习中的错误定位问题 reinforcement learning preference learning multimodal
23 Beyond Temporal Proximity: Explanatory World Models for Credit Assignment in Reinforcement Learning 提出解释性世界模型以解决强化学习中的信用分配问题 reinforcement learning PPO world model
24 Reinforcement Learning via Fixed-point Flow Matching 提出固定点流匹配方法以解决无数据强化学习问题 reinforcement learning flow matching multimodal
25 TabMF: Learning Dataset Representations with Tabular Foundation Models 提出TabMF框架以提升数据集表示学习效果 representation learning foundation model
26 JEPA-kMMVAE: CSI Prediction for Time-Varying Wireless Channels 提出JEPA-kMMVAE以解决时间变化无线信道的CSI预测问题 JEPA Joint-Embedding Predictive Architecture joint-embedding predictive architecture
27 Continually Post-training Small Models via Synthetic Self-distillation 提出合成自蒸馏方法以解决小模型持续学习问题 distillation instruction following
28 Goal-Oriented Bundle Management and Flow Control for Deep-Space Communications via Reinforcement Learning 提出基于强化学习的深空通信目标导向捆绑管理与流量控制方法 reinforcement learning PPO
29 ReplaySSM: Cache SSM Inputs, Not State 提出ReplaySSM以解决状态空间模型的瓶颈问题 Mamba SSM
30 Rethinking Relative Policy Differences Transfer of On-Policy Distillation 提出相对策略差异转移以优化在线蒸馏训练 reinforcement learning distillation
31 SafeDiSTL: Decomposing Joint Constraints for Safe Multi-Agent Reinforcement Learning 提出SafeDiSTL以解决多智能体强化学习中的联合约束问题 reinforcement learning
32 Understanding Mamba-3 Through Oscillatory Dynamics 提出Mamba-3以解决序列建模中的信息积累问题 Mamba
33 MORAD: Reinforcement Learning with Multi-Objective Rewards for RNA Inverse Design 提出MORAD框架以解决RNA逆向设计中的多目标优化问题 reinforcement learning
34 CAST: Credit Alignment for Cross-Tokenizer On-Policy Distillation 提出CAST以解决跨标记器蒸馏中的信用对齐问题 distillation
35 State-OPD: State-Grounded On-Policy Distillation for Multi-Turn Agents 提出State-OPD以解决多回合智能体的蒸馏训练问题 distillation
36 Personalized Federated Direct Preference Optimization via LoRA Adapters 提出个性化联邦直接偏好优化方法以解决异构偏好问题 direct preference optimization
37 AgentTune: A Unified Training Lifecycle for Tool-Using LLM Agents 提出AgentTune以统一工具使用代理的训练生命周期 reinforcement learning distillation ✅
38 Generalized Proximal Policy Optimization: Unifying LLM RL Algorithms from the Perspective of Divergence 提出广义近端策略优化以解决现有PPO方法的局限性 reinforcement learning PPO

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

#题目一句话要点标签🔗⭐
39 PPG-Bench: A Unified Benchmark for PPG Foundation Models 提出PPG-Bench以统一评估PPG基础模型的性能 PULSE foundation model ✅
40 PlacePulse: Learning Transferable Place Representations from Structural Geographic Data 提出PlacePulse以解决地理数据特征工程不足问题 PULSE foundation model

🔬 支柱一：机器人控制 (Robot Control) (1 篇)

#题目一句话要点标签🔗⭐
41 A Scalable Learned Iterative Solver for Nonlinear MPC via Temporal Decomposition 提出可扩展的学习迭代求解器以解决非线性模型预测控制问题 MPC model predictive control

🔬 支柱四：生成式动作 (Generative Motion) (1 篇)

#题目一句话要点标签🔗⭐
42 Opening the Black Box of Classifier-Free Guidance via Information Bottleneck 提出动态无分类器引导策略以优化扩散模型生成 classifier-free guidance

⬅️ 返回 cs.LG 首页 · 🏠 返回主页