cs.LG（2025-04-09）

📊 共 7 篇论文 | 🔗 1 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (4 🔗1) 支柱二：RL算法与架构 (RL & Architecture) (2) 支柱六：视频提取与匹配 (Video Extraction) (1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (4 篇)

#	题目	一句话要点	标签	🔗	⭐
1	ColonScopeX: Leveraging Explainable Expert Systems with Multimodal Data for Improved Early Diagnosis of Colorectal Cancer	ColonScopeX：利用可解释专家系统与多模态数据改进结直肠癌早期诊断	multimodal
2	A Neuro-inspired Interpretation of Unlearning in Large Language Models through Sample-level Unlearning Difficulty	提出基于神经科学启发的MRD指标，提升大语言模型中样本级别Unlearning的效率与效果。	large language model
3	Sculpting Subspaces: Constrained Full Fine-Tuning in LLMs for Continual Learning	提出基于自适应SVD的约束全参数微调方法，解决LLM持续学习中的灾难性遗忘问题。	large language model instruction following
4	Holistic Capability Preservation: Towards Compact Yet Comprehensive Reasoning Models	提出Ring-Lite-Distill：一种紧凑且全面的轻量级推理模型	large language model instruction following	✅

🔬 支柱二：RL算法与架构 (RL & Architecture) (2 篇)

#	题目	一句话要点	标签	🔗	⭐
5	Neural Motion Simulator: Pushing the Limit of World Models in Reinforcement Learning	提出神经运动模拟器MoSim，提升具身智能体在强化学习中的世界模型性能	reinforcement learning world model
6	Bridging the Gap Between Preference Alignment and Machine Unlearning	提出U2A框架，通过选择性遗忘负样本提升大语言模型偏好对齐性能	reinforcement learning RLHF large language model

🔬 支柱六：视频提取与匹配 (Video Extraction) (1 篇)

#	题目	一句话要点	标签	🔗	⭐
7	A Graph-Enhanced DeepONet Approach for Real-Time Estimating Hydrogen-Enriched Natural Gas Flow under Variable Operations	提出图增强DeepONet框架，用于实时估计氢气掺混天然气管道中的氢气浓度。	sparse sensors

⬅️ 返回 cs.LG 首页 · 🏠 返回主页