cs.AI（2026-07-21）

📊 共 23 篇论文 | 🔗 4 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (11 🔗4) 支柱二：RL算法与架构 (RL & Architecture) (9) 支柱七：动作重定向 (Motion Retargeting) (2) 支柱一：机器人控制 (Robot Control) (1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (11 篇)

#题目一句话要点标签🔗⭐
1 Quality Action Assurance: Multimodal Verification of Examiner Claims in VR OSCEs 提出质量行动保障框架以解决OSCE评分主观性问题 large language model multimodal
2 ResearchArena: Evaluating Sabotage and Monitoring in Automated AI R&D 提出ResearchArena框架以评估自动化AI研发中的破坏与监控问题 chain-of-thought
3 Assessment in Team Problem-Solving Exercises in Computing Education 提出基于聚类与大语言模型的团队评估方法以提升TTX反馈效率 large language model
4 SciCodePile: A 128GB Corpus and Executable Benchmark for Challenging Scientific Code Generation 提出SciCodePile以解决科学代码生成的挑战问题 large language model ✅
5 Mi-Memory: A Lifecycle Memory Framework for Personal AI 提出Mi-Memory框架以解决个人AI记忆管理问题 multimodal ✅
6 From Dependency to Compositionality: A Neurosymbolic Lifting of LLM Outputs via Combinatory Categorial Grammar 提出神经符号框架以提升LLM输出的组合性 large language model
7 PhoenixRepair: Rethinking Repair Strategy Exploration in Software Agents 提出PhoenixRepair以解决软件代理修复策略探索不足问题 large language model ✅
8 AI Tour Meeting: Group Travel Planning by LLM Agents 提出AI Tour Meeting框架以解决群体旅行规划问题 large language model
9 SkillSight: Seeing Through Shared Descriptions for Accurate Skill Retrieval 提出SkillSight以解决技能检索中的共享描述偏差问题 large language model ✅
10 SciHazard: A Benchmark for Measuring Scientific Safety Risks with Decomposed Harm Scoring 提出SciHazard基准以评估科学安全风险 large language model
11 CPInj: Uncovering Prompt Injection Risks in Textual Collaborative Prompt Optimization 提出CPInj以揭示文本协作提示优化中的注入风险 large language model

🔬 支柱二：RL算法与架构 (RL & Architecture) (9 篇)

#题目一句话要点标签🔗⭐
12 Strategy-Following Multi-Agent Deep Reinforcement Learning Considering Control Strategies Provided to Other Agents 提出一种多智能体深度强化学习方法以优化人类控制策略 reinforcement learning deep reinforcement learning
13 DWM: Separating World Effects from Actions in Latent World Models 提出DWM框架以解决潜在世界模型中的状态变化归因问题 world model world models
14 Do AI-Native Biotechs Need Departments? Benchmarking Company World Models for AI-Driven Drug Development 提出公司世界模型以优化AI驱动药物开发 world model world models
15 Fishing Out Free Riders: Shapley-Based Reward Attribution for Parallel Reasoning via Reinforcement Learning 提出Parallel Shapley框架以解决多路径推理中的奖励归属问题 reinforcement learning large language model
16 Comparative Study of Multi-Agent Actor-Critic Algorithms in Parameterized Action Reinforcement Learning 提出多智能体扩展的演员-评论家算法以提升参数化动作强化学习性能 reinforcement learning SAC
17 NaviAIS: A Scenario-Level Vessel Trajectory Prediction Dataset withVectorized Lane Priors and the NaviLane Forecasting Framework 提出NaviAIS数据集与NaviLane框架以解决船舶轨迹预测问题 world model world models multimodal
18 Measuring Reward-Seeking via Contrastive Belief Updates 提出对比合成文档微调以测量奖励寻求行为 reinforcement learning chain-of-thought
19 Athena-Brain Technical Report: An Efficient Robot Brain for General Intelligence and Embodied Interactio 提出Athena-Brain-8B以解决通用智能与具身交互的模型整合问题 reinforcement learning large language model
20 Black-Mamba: Biologically-Inspired Leaky Accumulation for Conceptual Knowledge under Distribution Drift 提出Black-Mamba以解决非平稳条件下的预测适应问题 Mamba

🔬 支柱七：动作重定向 (Motion Retargeting) (2 篇)

#题目一句话要点标签🔗⭐
21 Semantic Primes as Explanans for Emotion in Large Language Models 提出语义原语作为大语言模型情感解释的新方法 motion representation large language model
22 Enhancing Transformer-based Routing by Encoding Distance via Relative Positional Encoding 通过相对位置编码增强Transformer路由以解决团队定向问题 spatial relationship

🔬 支柱一：机器人控制 (Robot Control) (1 篇)

#题目一句话要点标签🔗⭐
23 Data Leakage Prevention in Agentic Applications via Preemptive Hardening 提出预防数据泄露的管道以解决多代理应用中的安全问题 manipulation

⬅️ 返回 cs.AI 首页 · 🏠 返回主页