cs.AI（2025-05-28）

📊 共 30 篇论文 | 🔗 3 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (22 🔗1) 支柱二：RL算法与架构 (RL & Architecture) (7 🔗2) 支柱六：视频提取与匹配 (Video Extraction) (1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (22 篇)

#题目一句话要点标签🔗⭐
1 Test-Time Immunization: A Universal Defense Framework Against Jailbreaks for (Multimodal) Large Language Models 提出Test-time Immunization (TIM)，用于防御(多模态)大语言模型的越狱攻击 large language model multimodal
2 What Makes a Good Reasoning Chain? Uncovering Structural Patterns in Long Chain-of-Thought Reasoning 提出LCoT2Tree框架，揭示并利用长链推理的结构化模式提升LLM性能 large language model chain-of-thought
3 Visual Large Language Models Exhibit Human-Level Cognitive Flexibility in the Wisconsin Card Sorting Test 研究表明视觉大语言模型在威斯康星卡片分类测试中展现出人类水平的认知灵活性 large language model chain-of-thought
4 Subspecialty-Specific Foundation Model for Intelligent Gastrointestinal Pathology Digepath：面向胃肠病理的专科基础模型，提升诊断精度与效率 foundation model
5 Permissioned LLMs: Enforcing Access Control in Large Language Models 提出Permissioned LLMs，在大型语言模型中强制执行访问控制，保障企业数据安全。 large language model
6 A Tool for Generating Exceptional Behavior Tests With Large Language Models exLong：利用大语言模型自动生成异常行为测试用例，提升软件测试覆盖率 large language model
7 A Large Language Model-Enabled Control Architecture for Dynamic Resource Capability Exploration in Multi-Agent Manufacturing Systems 提出基于大语言模型的控制架构，用于多智能体制造系统中动态资源能力探索 large language model
8 In Dialogue with Intelligence: Rethinking Large Language Models as Collective Knowledge 将大型语言模型视为集体知识，通过对话涌现智能 large language model
9 Effective and Efficient One-pass Compression of Speech Foundation Models Using Sparsity-aware Self-pinching Gates 提出一种基于稀疏感知的自收缩门控机制，用于高效压缩语音基础模型。 foundation model
10 From Large AI Models to Agentic AI: A Tutorial on Future Intelligent Communications 提出面向6G通信的LAMs和Agentic AI设计范式，解决智能通信系统中的感知、扩展和适应性挑战。 large language model multimodal
11 Generative Social Choice: The Next Generation 提出生成式社会选择框架扩展，解决语句长度预算约束下用户意见代表性摘要生成问题 large language model
12 Rethinking the Unsolvable: When In-Context Search Meets Test-Time Scaling 结合上下文搜索与测试时缩放，显著提升LLM在超难推理任务上的性能 large language model
13 Reinforced Reasoning for Embodied Planning 提出基于强化学习的推理框架，提升具身智能规划任务中的决策能力。 embodied AI
14 BugWhisperer: Fine-Tuning LLMs for SoC Hardware Vulnerability Detection BugWhisperer：微调LLM用于SoC硬件漏洞检测 large language model
15 Operationalizing CaMeL: Strengthening LLM Defenses for Enterprise Deployment 增强LLM防御能力：为企业部署优化CaMeL框架 large language model
16 Seven Security Challenges That Must be Solved in Cross-domain Multi-agent LLM Systems 揭示跨域多智能体LLM系统中七大安全挑战，为未来研究提供指导 large language model
17 AgentDNS: A Root Domain Naming System for LLM Agents AgentDNS：为LLM Agent设计的根域名命名系统，实现跨平台服务发现与安全调用。 large language model
18 ChatPD: An LLM-driven Paper-Dataset Networking System ChatPD：一个基于LLM的论文-数据集网络构建系统 large language model ✅
19 New Tools are Needed for Tracking Adherence to AI Model Behavioral Use Clauses 针对AI模型行为使用条款，提出追踪采纳与遵守情况的新工具需求 foundation model
20 Towards Efficient Key-Value Cache Management for Prefix Prefilling in LLM Inference 针对LLM推理中前缀预填充，提出高效的Key-Value缓存管理方案 large language model
21 Incorporating LLMs for Large-Scale Urban Complex Mobility Simulation 融合LLM的城市复杂交通大规模仿真，提升Agent多样性和真实性 large language model
22 Extracting Research Instruments from Educational Literature Using LLMs 利用大型语言模型从教育文献中提取研究工具信息 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🔗⭐
23 VIRAL: Vision-grounded Integration for Reward design And Learning VIRAL：基于视觉的奖励函数设计与学习框架，提升人机对齐 reinforcement learning reward design large language model ✅
24 Topological Structure Learning Should Be A Research Priority for LLM-Based Multi-Agent Systems 针对LLM多智能体系统，提出拓扑结构学习框架以优化智能体间交互 reinforcement learning large language model
25 Reward-Independent Messaging for Decentralized Multi-Agent Reinforcement Learning 提出MARL-CPC框架，实现去中心化多智能体强化学习中的奖励独立通信。 reinforcement learning
26 BinauralFlow: A Causal and Streamable Approach for High-Quality Binaural Speech Synthesis with Flow Matching Models BinauralFlow：基于流匹配模型的高质量、因果、流式双耳语音合成方法 flow matching
27 Improving Respiratory Sound Classification with Architecture-Agnostic Knowledge Distillation from Ensembles 利用架构无关的集成知识蒸馏提升呼吸音分类性能 distillation
28 Decomposing Elements of Problem Solving: What "Math" Does RL Teach? 分解问题解决要素：强化学习究竟教会了LLM什么“数学”？ reinforcement learning distillation ✅
29 HDDLGym: A Tool for Studying Multi-Agent Hierarchical Problems Defined in HDDL with OpenAI Gym HDDLGym：一个用于研究基于HDDL定义的多智能体分层问题的OpenAI Gym工具 reinforcement learning model-based RL

🔬 支柱六：视频提取与匹配 (Video Extraction) (1 篇)

#题目一句话要点标签🔗⭐
30 Efficiently Enhancing General Agents With Hierarchical-categorical Memory 提出EHC，一种无需参数更新即可有效增强通用Agent的分层分类记忆方法 HMR large language model

⬅️ 返回 cs.AI 首页 · 🏠 返回主页