cs.LG(2025-07-26)
📊 共 9 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) (4 🔗1)
支柱一:机器人控制 (Robot Control) (2)
支柱八:物理动画 (Physics-based Animation) (2)
支柱九:具身大模型 (Embodied Foundation Models) (1)
🔬 支柱二:RL算法与架构 (RL & Architecture) (4 篇)
| # | 题目 | 一句话要点 | 标签 | 🔗 | ⭐ |
|---|---|---|---|---|---|
| 1 | Inducing Causal World Models in LLMs for Zero-Shot Physical Reasoning | 提出CWMI框架,通过在LLM中嵌入因果世界模型实现零样本物理推理 | world model large language model multimodal | ||
| 2 | CANDLE: A Cross-Modal Agentic Knowledge Distillation Framework for Interpretable Sarcopenia Diagnosis | CANDLE:一种用于可解释性肌少症诊断的跨模态Agent知识蒸馏框架 | reinforcement learning distillation large language model | ||
| 3 | GNSP: Gradient Null Space Projection for Preserving Cross-Modal Alignment in VLMs Continual Learning | 提出GNSP方法,通过梯度零空间投影和模态对齐保持,解决VLM持续学习中的灾难性遗忘问题。 | distillation multimodal | ||
| 4 | Agentic Reinforced Policy Optimization | 提出Agentic Reinforced Policy Optimization (ARPO)以提升LLM在多轮工具交互推理中的性能。 | reinforcement learning large language model | ✅ |
🔬 支柱一:机器人控制 (Robot Control) (2 篇)
| # | 题目 | 一句话要点 | 标签 | 🔗 | ⭐ |
|---|---|---|---|---|---|
| 5 | VAE-GAN Based Price Manipulation in Coordinated Local Energy Markets | 提出基于VAE-GAN的价格操纵策略,用于评估本地能源市场中协调机制的鲁棒性。 | manipulation reinforcement learning | ||
| 6 | Strategic Filtering for Content Moderation: Free Speech or Free of Distortion? | 提出一种面向内容审核的策略性过滤方法,平衡言论自由与失真控制。 | manipulation |
🔬 支柱八:物理动画 (Physics-based Animation) (2 篇)
| # | 题目 | 一句话要点 | 标签 | 🔗 | ⭐ |
|---|---|---|---|---|---|
| 7 | Quantum-Informed Machine Learning for Predicting Spatiotemporal Chaos | 提出量子信息机器学习框架,用于预测高维时空混沌系统的长期动态行为。 | spatiotemporal | ||
| 8 | Sparse-mode Dynamic Mode Decomposition for Disambiguating Local and Global Structures | 提出稀疏模式动态模态分解,用于区分局部和全局结构 | spatiotemporal |
🔬 支柱九:具身大模型 (Embodied Foundation Models) (1 篇)
| # | 题目 | 一句话要点 | 标签 | 🔗 | ⭐ |
|---|---|---|---|---|---|
| 9 | Large Language Model Agent for Structural Drawing Generation Using ReAct Prompt Engineering and Retrieval Augmented Generation | 提出基于LLM Agent的结构图生成方法,结合ReAct提示工程和RAG提升绘图质量。 | large language model |