cs.LG（2025-04-23）

📊 共 7 篇论文 | 🔗 1 篇有代码

🎯 兴趣领域导航

支柱二：RL算法与架构 (RL & Architecture) (3) 支柱九：具身大模型 (Embodied Foundation Models) (2 🔗1) 支柱一：机器人控制 (Robot Control) (1) 支柱四：生成式动作 (Generative Motion) (1)

🔬 支柱二：RL算法与架构 (RL & Architecture) (3 篇)

#	题目	一句话要点	标签	🔗	⭐
1	(Im)possibility of Automated Hallucination Detection in Large Language Models	从理论上分析大语言模型幻觉检测的（不）可能性	reinforcement learning RLHF large language model
2	I-Con: A Unifying Framework for Representation Learning	I-Con：一个统一的表征学习框架，通过最小化积分KL散度泛化多种损失函数。	representation learning contrastive learning
3	Representation Learning via Non-Contrastive Mutual Information	提出MINC非对比互信息最大化方法，提升自监督表征学习效果。	representation learning

🔬 支柱九：具身大模型 (Embodied Foundation Models) (2 篇)

#	题目	一句话要点	标签	🔗	⭐
4	ParetoHqD: Fast Offline Multiobjective Alignment of Large Language Models using Pareto High-quality Data	ParetoHqD：利用Pareto高质量数据快速离线多目标对齐大型语言模型	large language model
5	Process Reward Models That Think	提出ThinkPRM：一种数据高效的生成式过程奖励模型，用于提升测试时推理能力。	chain-of-thought	✅

🔬 支柱一：机器人控制 (Robot Control) (1 篇)

#	题目	一句话要点	标签	🔗	⭐
6	PIN-WM: Learning Physics-INformed World Models for Non-Prehensile Manipulation	提出PIN-WM，用于学习非抓取操作的物理信息世界模型，实现Sim2Real迁移。	manipulation sim2real real2sim

🔬 支柱四：生成式动作 (Generative Motion) (1 篇)

#	题目	一句话要点	标签	🔗	⭐
7	OccuEMBED: Occupancy Extraction Merged with Building Energy Disaggregation for Occupant-Responsive Operation at Scale	OccuEMBED：融合占用提取与建筑能耗分解，实现大规模的居住者响应式建筑运行	penetration

⬅️ 返回 cs.LG 首页 · 🏠 返回主页