cs.AI（2025-05-20）

📊 共 44 篇论文 | 🔗 5 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (30 🔗3) 支柱二：RL算法与架构 (RL & Architecture) (10 🔗1) 支柱一：机器人控制 (Robot Control) (3) 支柱八：物理动画 (Physics-based Animation) (1 🔗1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (30 篇)

#题目一句话要点标签🔗⭐
1 Large Language Model-Driven Distributed Integrated Multimodal Sensing and Semantic Communications 提出LLM-DiSAC框架以解决单模态感知系统的局限性 large language model multimodal
2 Multimodal RAG-driven Anomaly Detection and Classification in Laser Powder Bed Fusion using Large Language Models 提出多模态RAG驱动框架以解决激光粉末床熔融中的异常检测问题 large language model multimodal
3 Towards a Foundation Model for Communication Systems 提出一种基础模型以解决通信系统中的多模态数据处理问题 foundation model multimodal
4 Debating for Better Reasoning: An Unsupervised Multimodal Approach 提出多模态辩论框架以提升视觉问答性能 large language model multimodal
5 Large Language Model Powered Decision Support for a Metal Additive Manufacturing Knowledge Graph 提出金属增材制造知识图谱与大语言模型结合的决策支持系统 large language model
6 SAFEPATH: Preventing Harmful Reasoning in Chain-of-Thought via Early Alignment 提出SAFEPATH以解决大型推理模型的安全性问题 chain-of-thought
7 Can Large Language Models Really Recognize Your Name? 提出AMBENCH基准以解决LLM隐私识别问题 large language model
8 Guarded Query Routing for Large Language Models 提出受保护的查询路由方法以解决大语言模型的查询分类问题 large language model ✅
9 Toward Embodied AGI: A Review of Embodied AI and the Road Ahead 提出系统分类以推动具身人工智能的发展 embodied AI
10 Multimodal Mixture of Low-Rank Experts for Sentiment Analysis and Emotion Recognition 提出多模态低秩专家混合模型以解决情感分析和情绪识别问题 multimodal
11 The Multimodal Information Based Speech Processing (MISP) 2025 Challenge: Audio-Visual Diarization and Recognition 提出多模态会议转录方法以解决复杂声学条件下的挑战 multimodal
12 Agent Context Protocols Enhance Collective Inference 提出Agent上下文协议以增强多智能体集体推理能力 generalist agent multimodal
13 MLZero: A Multi-Agent System for End-to-end Machine Learning Automation 提出MLZero以实现多模态数据的端到端机器学习自动化 large language model multimodal
14 Reasoning Models Better Express Their Confidence 提出推理模型以提高信心表达的准确性 large language model chain-of-thought
15 ProMind-LLM: Proactive Mental Health Care via Causal Reasoning with Sensor Data 提出ProMind-LLM以解决心理健康评估中的主观性问题 large language model chain-of-thought
16 DrugPilot: LLM-based Parameterized Reasoning Agent for Drug Discovery 提出DrugPilot以解决药物发现中的多模态数据处理问题 large language model multimodal
17 Towards Embodied Cognition in Robots via Spatially Grounded Synthetic Worlds 提出空间基础合成世界以促进机器人具身认知 embodied AI
18 Two Experts Are All You Need for Steering Thinking: Reinforcing Cognitive Effort in MoE Reasoning Models Without Additional Training 提出RICE方法以解决MoE推理模型中的认知效率问题 instruction following
19 JARVIS: A Multi-Agent Code Assistant for High-Quality EDA Script Generation 提出JARVIS框架以解决EDA脚本生成质量问题 large language model
20 SATBench: Benchmarking LLMs' Logical Reasoning via Automated Puzzle Generation from SAT Formulas 提出SATBench以评估大型语言模型的逻辑推理能力 large language model ✅
21 Balanced and Elastic End-to-end Training of Dynamic LLMs 提出DynMo以解决大规模动态LLM训练中的负载不均问题 large language model
22 ContextAgent: Context-Aware Proactive LLM Agents with Open-World Sensory Perceptions 提出ContextAgent以解决现有主动智能体的局限性问题 large language model ✅
23 From nuclear safety to LLM security: Applying non-probabilistic risk management strategies to build safe and secure LLM-powered systems 提出非概率风险管理策略以解决LLM安全问题 large language model
24 Towards Reliable Proof Generation with LLMs: A Neuro-Symbolic Approach 提出神经符号方法以解决数学证明生成问题 large language model
25 Choosing a Model, Shaping a Future: Comparing LLM Perspectives on Sustainability and its Relationship with AI 比较五种大型语言模型对可持续性与AI关系的看法 large language model
26 Knowledge Graph Based Repository-Level Code Generation 提出基于知识图谱的代码生成方法以提升代码检索质量 large language model
27 SCAN: Semantic Document Layout Analysis for Textual and Visual Retrieval-Augmented Generation 提出SCAN以解决丰富文档的检索增强生成问题 large language model
28 Divide by Question, Conquer by Agent: SPLIT-RAG with Question-Driven Graph Partitioning 提出SPLIT-RAG以解决大规模知识图谱的检索效率与准确性问题 large language model
29 RAG/LLM Augmented Switching Driven Polymorphic Metaheuristic Framework 提出自适应多态元启发式框架以解决优化问题 multimodal
30 LLM-based Evaluation Policy Extraction for Ecological Modeling 提出基于LLM的评估策略提取以解决生态建模评估问题 large language model

🔬 支柱二：RL算法与架构 (RL & Architecture) (10 篇)

#题目一句话要点标签🔗⭐
31 Reinforcement Learning from User Feedback 提出用户反馈强化学习框架以解决用户偏好对齐问题 reinforcement learning RLHF large language model
32 Reinforcement Learning vs. Distillation: Understanding Accuracy and Capability in LLM Reasoning 探讨RLVR与蒸馏在LLM推理中的准确性与能力差异 reinforcement learning distillation
33 Causal Cartographer: From Mapping to Reasoning Over Counterfactual Worlds 提出Causal Cartographer以解决因果推理与反事实评估问题 world model large language model foundation model
34 DSMentor: Enhancing Data Science Agents with Curriculum Learning and Online Knowledge Accumulation 提出DSMentor以优化数据科学代理的推理过程 curriculum learning large language model
35 SHARP: Synthesizing High-quality Aligned Reasoning Problems for Large Reasoning Models Reinforcement Learning 提出SHARP以解决大规模推理模型训练中的问题生成挑战 reinforcement learning chain-of-thought
36 RL of Thoughts: Navigating LLM Reasoning with Inference-time Reinforcement Learning 提出RL-of-Thoughts以增强大语言模型推理能力 reinforcement learning large language model
37 Visual Instruction Bottleneck Tuning 提出视觉指令瓶颈调优以提升多模态大语言模型的鲁棒性 representation learning large language model multimodal
38 Embedded Mean Field Reinforcement Learning for Perimeter-defense Game 提出嵌入式均场强化学习框架以解决复杂的周边防御游戏问题 reinforcement learning representation learning
39 PRL: Prompts from Reinforcement Learning 提出基于强化学习的自动提示生成方法PRL以解决提示工程挑战 reinforcement learning ✅
40 TelePlanNet: An AI-Driven Framework for Efficient Telecom Network Planning 提出TelePlanNet以解决5G网络基站选址优化问题 reinforcement learning large language model

🔬 支柱一：机器人控制 (Robot Control) (3 篇)

#题目一句话要点标签🔗⭐
41 Self-Evolving Curriculum for LLM Reasoning 提出自演化课程以优化大语言模型推理能力 dual-arm reinforcement learning curriculum learning
42 EVA: Red-Teaming GUI Agents via Evolving Indirect Prompt Injection 提出EVA框架以应对间接提示注入攻击问题 manipulation multimodal
43 Transductively Informed Inductive Program Synthesis 提出TIIPS框架以提升程序合成的准确性与泛化能力 manipulation

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

#题目一句话要点标签🔗⭐
44 AudioJailbreak: Jailbreak Attacks against End-to-End Large Audio-Language Models 提出AudioJailbreak以解决音频语言模型的安全漏洞问题 PULSE ✅

⬅️ 返回 cs.AI 首页 · 🏠 返回主页