cs.AI(2025-02-11)
📊 共 9 篇论文
🎯 兴趣领域导航
支柱九:具身大模型 (Embodied Foundation Models) (5)
支柱二:RL算法与架构 (RL & Architecture) (2)
支柱一:机器人控制 (Robot Control) (2)
🔬 支柱九:具身大模型 (Embodied Foundation Models) (5 篇)
| # | 题目 | 一句话要点 | 标签 | 🔗 | ⭐ |
|---|---|---|---|---|---|
| 1 | Universal Adversarial Attack on Aligned Multimodal LLMs | 提出针对多模态LLM的通用对抗攻击,利用单张优化图像绕过对齐安全措施。 | large language model multimodal | ||
| 2 | When Incentives Backfire, Data Stops Being Human | 重新思考数据收集系统,利用内在动机维持高质量数据来源 | large language model | ||
| 3 | Deep Semantic Graph Learning via LLM based Node Enhancement | 提出基于LLM增强节点表示的深度语义图学习框架,提升节点分类性能 | large language model | ||
| 4 | From Hazard Identification to Controller Design: Proactive and LLM-Supported Safety Engineering for ML-Powered Systems | 提出LLM辅助的主动安全工程方法,解决ML系统潜在风险识别与控制问题 | large language model | ||
| 5 | Trustworthy AI: Safety, Bias, and Privacy -- A Survey | 针对AI系统安全性、偏见和隐私问题,提出可信AI的综合性调研。 | large language model |
🔬 支柱二:RL算法与架构 (RL & Architecture) (2 篇)
| # | 题目 | 一句话要点 | 标签 | 🔗 | ⭐ |
|---|---|---|---|---|---|
| 6 | When More is Less: Understanding Chain-of-Thought Length in LLMs | 揭示LLM中思维链长度与性能的非单调关系,并提出自适应CoT校准方法 | reinforcement learning large language model chain-of-thought | ||
| 7 | Polynomial-Time Approximability of Constrained Reinforcement Learning | 针对约束马尔可夫决策过程,提出多项式时间近似算法,解决多种约束下的策略优化问题。 | reinforcement learning |
🔬 支柱一:机器人控制 (Robot Control) (2 篇)
| # | 题目 | 一句话要点 | 标签 | 🔗 | ⭐ |
|---|---|---|---|---|---|
| 8 | ImitDiff: Transferring Foundation-Model Priors for Distraction Robust Visuomotor Policy | ImitDiff:利用预训练模型先验知识,提升视觉运动策略在复杂场景下的鲁棒性 | manipulation imitation learning foundation model | ||
| 9 | Human Decision-making is Susceptible to AI-driven Manipulation | 研究表明人类决策易受AI驱动的操纵,尤其在金融和情感决策中 | manipulation |