cs.LG（2025-12-01）
📊 共 7 篇论文
🎯 兴趣领域导航
🔬 支柱九：具身大模型 (Embodied Foundation Models) (4 篇)
| # | 题目 | 一句话要点 | 标签 | 🔗 | ⭐ |
|---|---|---|---|---|---|
| 1 | Do Large Language Models Walk Their Talk? Measuring the Gap Between Implicit Associations, Self-Report, and Behavioral Altruism | 评估大型语言模型利他行为：揭示内隐认知、自我报告与实际行为间的差距 | large language model | ||
| 2 | RE-LLM: Integrating Large Language Models into Renewable Energy Systems | RE-LLM：集成大语言模型到可再生能源系统，提升能源模型可解释性 | large language model | ||
| 3 | AlignSAE: Concept-Aligned Sparse Autoencoders | 提出AlignSAE，通过概念对齐的稀疏自编码器实现LLM内部知识的可控干预。 | large language model | ||
| 4 | Zero-Overhead Introspection for Adaptive Test-Time Compute | ZIP-RC：为LLM配备零开销自省能力，实现自适应测试时计算。 | large language model |
🔬 支柱二：RL算法与架构 (RL & Architecture) (3 篇)
| # | 题目 | 一句话要点 | 标签 | 🔗 | ⭐ |
|---|---|---|---|---|---|
| 5 | Forecasting in Offline Reinforcement Learning for Non-stationary Environments | 提出FORL框架，解决离线强化学习在非平稳环境中因状态偏移导致的性能下降问题。 | reinforcement learning offline RL offline reinforcement learning | ||
| 6 | Stabilizing Reinforcement Learning with LLMs: Formulation and Practices | 提出基于LLM的强化学习新公式，解决训练不稳定问题并提供稳定训练方案。 | reinforcement learning large language model | ||
| 7 | Agentic Policy Optimization via Instruction-Policy Co-Evolution | 提出INSPO，通过指令-策略协同进化优化Agentic策略，提升多轮推理能力。 | reinforcement learning large language model |