cs.AI（2026-03-20）

📊 共 17 篇论文 | 🔗 2 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (10 🔗1) 支柱二：RL算法与架构 (RL & Architecture) (5 🔗1) 支柱一：机器人控制 (Robot Control) (1) 支柱八：物理动画 (Physics-based Animation) (1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (10 篇)

#题目一句话要点标签🔗⭐
1 Evolving Jailbreaks: Automated Multi-Objective Long-Tail Attacks on Large Language Models 提出EvoJail框架以自动化发现长尾攻击策略 large language model
2 Embodied Science: Closing the Discovery Loop with Agentic Embodied AI 提出具身科学范式，利用具身AI闭环解决科学发现难题 embodied AI
3 AI Agents Can Already Autonomously Perform Experimental High Energy Physics AI Agent自主执行高能物理实验分析，加速科研流程 large language model
4 Learning Dynamic Belief Graphs for Theory-of-mind Reasoning 提出动态信念图模型，增强LLM在复杂环境中基于心理理论的推理能力 large language model
5 Pitfalls in Evaluating Interpretability Agents 提出无监督内在评估方法以解决自动可解释性系统评估挑战 large language model
6 Agentic Harness for Real-World Compilers 提出llvm-autofix，用于辅助LLM智能体理解和修复LLVM编译器漏洞。 large language model ✅
7 Utility-Guided Agent Orchestration for Efficient LLM Tool Use 提出效用引导的代理编排以优化LLM工具使用效率 large language model
8 Stepwise: Neuro-Symbolic Proof Search for Automated Systems Verification 提出Stepwise神经符号框架，用于自动化系统验证中的定理证明搜索。 large language model
9 Skilled AI Agents for Embedded and IoT Systems Development 提出基于技能的AI Agent框架，用于硬件在环嵌入式和物联网系统开发 large language model
10 Optimal Scalar Quantization for Matrix Multiplication: Closed-Form Density and Phase Transition 提出最佳标量量化方法以优化矩阵乘法精度 large language model

🔬 支柱二：RL算法与架构 (RL & Architecture) (5 篇)

#题目一句话要点标签🔗⭐
11 Experience is the Best Teacher: Motivating Effective Exploration in Reinforcement Learning for LLMs 提出HeRL框架，通过回溯经验引导LLM强化学习中的有效探索 reinforcement learning large language model ✅
12 A Subgoal-driven Framework for Improving Long-Horizon LLM Agents 提出子目标驱动框架，提升LLM Agent在长程任务中的性能 reinforcement learning large language model
13 PA2D-MORL: Pareto Ascent Directional Decomposition based Multi-Objective Reinforcement Learning 提出基于帕累托上升方向分解的多目标强化学习方法，解决复杂机器人控制问题 reinforcement learning
14 Improving Generalization on Cybersecurity Tasks with Multi-Modal Contrastive Learning 提出多模态对比学习框架，提升网络安全任务中的泛化能力 contrastive learning
15 PolicySim: An LLM-Based Agent Social Simulation Sandbox for Proactive Policy Optimization PolicySim：基于LLM的社交模拟沙箱，用于主动优化平台策略 DPO direct preference optimization

🔬 支柱一：机器人控制 (Robot Control) (1 篇)

#题目一句话要点标签🔗⭐
16 Trojan's Whisper: Stealthy Manipulation of OpenClaw through Injected Bootstrapped Guidance 通过注入引导指令隐蔽操纵OpenClaw：一种新型的自主编码代理攻击方法 manipulation

🔬 支柱八：物理动画 (Physics-based Animation) (1 篇)

#题目一句话要点标签🔗⭐
17 Offshore oil and gas platform dynamics in the North Sea, Gulf of Mexico, and Persian Gulf: Exploiting the Sentinel-1 archive 利用Sentinel-1数据和深度学习自动监测三大海域油气平台时空动态 spatiotemporal

⬅️ 返回 cs.AI 首页 · 🏠 返回主页