cs.AI（2026-08-20）

📊 共 28 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) (16 🔗2) 支柱二：RL算法与架构 (RL & Architecture) (7) 支柱一：机器人控制 (Robot Control) (4) 支柱八：物理动画 (Physics-based Animation) (1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (16 篇)

#	题目	一句话要点	标签	🔗	⭐
1	Rule-Compliant Visual Spatial Planning for Multimodal Large Language Models	提出RuleMaze以解决多模态大语言模型的空间规划问题	large language model multimodal	✅
2	Towards general embodied intelligence: integrating large language models, knowledge bases, and reasoning capabilities to build the next generation of AI agents	提出整合大语言模型与知识库以推进通用体现智能	large language model multimodal
3	EchoCoT: Extracting Hidden Chain-of-Thought from Large Reasoning Models	提出EchoCoT以提取大型推理模型中的隐含思维链	chain-of-thought
4	Frequency-Aware Continual Learning for Smart Contract Vulnerability Detection with Large Language Models	提出频率感知的持续学习方法以解决智能合约漏洞检测问题	large language model
5	MemTrapBench: Benchmarking Cognitive Traps in LLM Memory Use	提出MemTrapBench以评估大语言模型中的认知陷阱	large language model
6	Break It Down, Pass It On: Cross-Task Skill Transfer in LLM Agents	提出跨任务技能转移方法以提升LLM代理的能力	large language model
7	Optimal Skill Selection for LLM Agents with Provable Bicriteria Guarantees	提出最佳前缀选择算法以优化LLM代理的技能选择	large language model
8	Repo0: Design-Driven Zero-to-All Code Generation	提出Repo0框架以解决零到全代码生成问题	large language model
9	LLMs as Acquisition Policies for Finite-Pool Materials Optimization: A Controlled Study	利用开放权重的大语言模型优化有限材料池的获取策略	large language model
10	Volumetric Radiology AI in the Era of Multimodal Large Language Models	提出多模态大语言模型以解决体积放射学AI的表示不匹配问题	large language model foundation model multimodal
11	AEGIS: Preventing Cross-Domain Resource Abuse in MCP	提出AEGIS以解决MCP中的跨域资源滥用问题	large language model multimodal
12	Large Scale AI Grading of Handwritten Physics Assessments: Score Agreement and Olympiad Team Selection Outcomes	基于GPT-5.5的手写物理评估AI评分系统提升评分一致性	multimodal
13	FL-MAESTRO: Multi-Agent LLM Orchestration for Resource-Constrained Federated Learning	提出FL-MAESTRO以解决资源受限的联邦学习中的决策问题	large language model	✅
14	Terminal Agents: A Survey of AI Agents in Command-Line Environments	提出终端智能体框架以整合命令行环境中的AI行为研究	large language model
15	Peer-Voted LLM-Agent Stress Tests Find Feed-Induced Lexical Convergence but No Reliable Matched-Exposure Advantage for Distributed Sources	提出PV-SST以评估LLM代理的词汇收敛性	large language model
16	An LLM agent for end-to-end computational materials discovery	提出MAESTRO以解决计算材料发现中的多尺度任务协调问题	large language model

🔬 支柱二：RL算法与架构 (RL & Architecture) (7 篇)

#	题目	一句话要点	标签	🔗	⭐
17	Contrastive Mixed Prompt Learning for Incomplete Multimodal Sentiment Analysis with Unseen Modality Combination	提出对比混合提示学习以解决未见模态组合的多模态情感分析问题	contrastive learning multimodal
18	ADAPT: Physics-Aware Diffusion-based World Models for Adaptive Predictive Transferable HVAC Control	提出ADAPT以解决HVAC控制中的能耗与舒适度问题	reinforcement learning world model world models
19	An Irreducible Quantum Advantage in Aligning World Models with Reality	提出量子世界模型以解决经典模型无法对齐现实的问题	world model world models
20	SAPO: Single-Rollout Autoregressive Policy Optimization for Agentic Reinforcement Learning	提出SAPO以解决长时间交互任务中的策略优化问题	reinforcement learning PPO large language model
21	An Agentic Approach for Active Data Collection, Travel Behavior Modeling, and Weather-Sensitive Demand Prediction	提出三代理工作流程以优化数据收集与需求预测	predictive model large language model multimodal
22	MidTool: Mid-training Data Synthesis for Agentic Tool Use	提出MidTool以增强大型语言模型的工具使用能力	reinforcement learning affordance large language model
23	Enforcing LLM Safety through DMD-based Classification of Prompt-Response Embedding Dynamics	通过DMD分类提升大型语言模型的安全性	predictive model large language model

🔬 支柱一：机器人控制 (Robot Control) (4 篇)

#	题目	一句话要点	标签	🔗	⭐
24	EXIMO: VLM Guided Exploration of VLA Policies	提出EXIMO以解决机器人策略微调效率低下问题	manipulation teleoperation reinforcement learning
25	DECOWAM: Decoupled Whole-Body World-Action Model for Legged Mobile Manipulation	提出DECOWAM以解决移动操控中的视觉预测问题	whole-body control locomotion manipulation
26	Learning Hierarchical Skill Policies with Offline Quality-Diversity Reinforcement Learning	提出QDOS以解决离线强化学习中的技能提取问题	locomotion manipulation reinforcement learning
27	DECOWAM: Decoupled Whole-Body World-Action Model for Legged Mobile Manipulation	提出DECOWAM以解决移动操控中的视觉预测问题	whole-body control locomotion manipulation

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

#	题目	一句话要点	标签	🔗	⭐
28	STCO: Conditional Neural Operators for Time-Dependent PDEs	提出STCO以解决时间依赖PDEs的条件预测问题	spatiotemporal

⬅️ 返回 cs.AI 首页 · 🏠 返回主页