cs.AI（2025-12-27）

📊 共 11 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) (7) 支柱二：RL算法与架构 (RL & Architecture) (2 🔗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) (7 篇)

#题目一句话要点标签🔗⭐
1 Lessons from Neuroscience for AI: How integrating Actions, Compositional Structure and Episodic Memory could enable Safe, Interpretable and Human-Like AI 融合动作、组合结构与情景记忆，提升AI安全性、可解释性和类人能力 large language model foundation model chain-of-thought
2 Nightjar: Dynamic Adaptive Speculative Decoding for Large Language Models Serving Nightjar：一种动态自适应推测解码方法，提升大语言模型服务吞吐量。 large language model
3 Learning Multi-Modal Mobility Dynamics for Generalized Next Location Recommendation 提出M³ob模型，利用多模态时空知识增强下一位置推荐的泛化能力。 large language model
4 TravelBench: A Broader Real-World Benchmark for Multi-Turn and Tool-Using Travel Planning 提出TravelBench：一个更广泛的真实世界旅行规划多轮对话与工具使用基准 large language model
5 The Wisdom of Deliberating AI Crowds: Does Deliberation Improve LLM-Based Forecasting? 通过群体审议提升LLM预测能力：一种基于LLM间互相审查的改进方法 large language model
6 Hierarchical Pedagogical Oversight: A Multi-Agent Adversarial Framework for Reliable AI Tutoring 提出分层教学监督框架，利用对抗性多智能体提升AI辅导的可靠性 large language model
7 Monadic Context Engineering 提出Monadic Context Engineering，为自主Agent设计提供形式化基础。 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🔗⭐
8 Memento 2: Learning by Stateful Reflective Memory 提出基于状态反射记忆的Memento 2，用于大型语言模型智能体的持续和经验学习。 reinforcement learning large language model
9 RollArt: Scaling Agentic RL Training via Disaggregated Infrastructure RollArc：通过分离式基础设施扩展Agentic RL训练，提升训练吞吐量。 reinforcement learning large language model ✅

🔬 支柱一：机器人控制 (Robot Control) (1 篇)

#题目一句话要点标签🔗⭐
10 DarkPatterns-LLM: A Multi-Layer Benchmark for Detecting Manipulative and Harmful AI Behavior 提出DarkPatterns-LLM以解决AI行为操控检测问题 manipulation large language model

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

#题目一句话要点标签🔗⭐
11 Multi-AI Agent Framework Reveals the "Oxide Gatekeeper" in Aluminum Nanoparticle Oxidation 多智能体AI框架揭示铝纳米颗粒氧化过程中的“氧化物守门人”机制 spatiotemporal

⬅️ 返回 cs.AI 首页 · 🏠 返回主页