cs.LG（2026-06-05）

📊 共 23 篇论文 | 🔗 2 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (11 🔗2) 支柱二：RL算法与架构 (RL & Architecture) (10) 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) (11 篇)

#题目一句话要点标签🔗⭐
1 A robust PPG foundation model using multimodal physiological supervision 提出一种鲁棒的PPG基础模型以解决数据泛化问题 foundation model multimodal
2 GlucoFM-Bench: Benchmarking Time-Series Foundation Models for Blood Glucose Forecasting 提出GlucoFM-Bench以评估血糖预测的时间序列基础模型 large language model foundation model
3 TabSwift: An Efficient Tabular Foundation Model with Row-Wise Attention 提出TabSwift以提升表格基础模型的推理效率 foundation model
4 TALAN: Task-Aligned Latent Adaptation Networks for Targeted Post-Training of Large Language Models 提出TALAN以解决大语言模型后训练的针对性适应问题 large language model
5 Federated Foundation Models over Vehicular Networks 提出多模态多任务联邦基础模型以提升车载网络智能 foundation model ✅
6 Closed-Form Spectral Regularization for Multi-Task Model Merging 提出闭式谱正则化以解决多任务模型合并问题 foundation model multimodal
7 Sparse Subspace-to-Expert Sharing for Task-Agnostic Continual Learning 提出SETA框架以解决大语言模型的持续学习问题 large language model
8 A Comprehensive Anatomy of Human and DeepSeek-R1 LLM Mathematical Reasoning 揭示DeepSeek-R1在数学推理中的局限性与改进方向 large language model
9 DEFINED: A Data-Efficient Computational Framework for Fine-Grained Creativity Assessment in Debate Scenarios 提出DEFINED框架以解决辩论场景中的创造力评估问题 large language model
10 OffQ: Taming Structured Outliers in LLM Quantization by Offsetting 提出OffQ以解决LLM量化中的激活异常问题 large language model
11 RASFT: Rollout-Adaptive Supervised Fine-Tuning for Reasoning 提出RASFT以解决推理任务中的过拟合问题 large language model ✅

🔬 支柱二：RL算法与架构 (RL & Architecture) (10 篇)

#题目一句话要点标签🔗⭐
12 Bootstrap Theory of Representational Emergence: Explanatory Insufficiency as a Driver of Representation Learning and World Models 提出代表性出现的引导理论以解决现有表示不足问题 world model world models representation learning
13 CF-JEPA: Mask-free forward prediction with asymmetric encoder utilization for time-series representation learning 提出CF-JEPA以解决时间序列表示学习中的掩蔽问题 JEPA Joint-Embedding Predictive Architecture joint-embedding predictive architecture
14 Time series Foundation Models based on Physics-Informed Synthetic Histories for Cold-Start Photovoltaic Forecasting 提出基于物理信息合成历史的时间序列基础模型以解决冷启动光伏预测问题 MAE foundation model
15 Learning Explicit Behavioral Models with Adaptive Questions and World-Model Probes 提出显式符号行为模型以解决交互代理适应性不足问题 policy learning world model world models
16 On the Geometry of On-Policy Distillation 提出一种新方法以理解在政策蒸馏中的参数更新几何特性 reinforcement learning distillation large language model
17 SCALE: Scalable Cross-Attention Learning with Extrapolation for Agentic Workflow Scheduling 提出SCALE以解决异构集群调度的可扩展性问题 reinforcement learning deep reinforcement learning DRL
18 Self-evolving LLM agents with in-distribution Optimization 提出Q-Evolve框架以解决长时决策中的信用分配问题 reinforcement learning policy learning large language model
19 Unsupervised Continual Clustering via Forward-Backward Knowledge Distillation 提出无监督持续聚类方法以解决灾难性遗忘问题 distillation
20 A Held-Out Transition-Pair Falsifier for Long-Horizon Non-Abelian State Tracking 提出持出转换对伪造器以解决长时间非阿贝尔状态跟踪问题 SSM OMOMO
21 SlimSearcher: Training Efficiency-Aware Web Agents via Adaptive Reward Gating 提出SlimSearcher以解决深度研究代理的计算效率问题 reinforcement learning reward shaping

🔬 支柱一：机器人控制 (Robot Control) (1 篇)

#题目一句话要点标签🔗⭐
22 GenPO++: Generative Policy Optimization with Jacobian-free Likelihood Ratios 提出GenPO++以解决生成策略在强化学习中的概率评估问题 manipulation reinforcement learning policy learning

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

#题目一句话要点标签🔗⭐
23 Unified Geometry-Guided ML-FTLE for Tracking Transient Chaos from Scalar Time Series 提出几何引导的机器学习框架以追踪标量时间序列中的瞬态混沌 occupancy grid

⬅️ 返回 cs.LG 首页 · 🏠 返回主页