cs.AI（2025-05-27）

📊 共 46 篇论文 | 🔗 6 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (29 🔗3) 支柱二：RL算法与架构 (RL & Architecture) (13 🔗3) 支柱三：空间感知与语义 (Perception & Semantics) (2) 支柱一：机器人控制 (Robot Control) (2)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (29 篇)

#题目一句话要点标签🔗⭐
1 MSEarth: A Multimodal Scientific Dataset and Benchmark for Phenomena Uncovering in Earth Science 提出MSEarth：一个用于地球科学现象理解的多模态科学数据集与基准。 large language model multimodal
2 Privacy-Preserving Chest X-ray Report Generation via Multimodal Federated Learning with ViT and GPT-2 提出基于ViT和GPT-2的多模态联邦学习框架，用于保护隐私的胸部X光报告生成。 multimodal
3 WDMIR: Wavelet-Driven Multimodal Intent Recognition 提出WDMIR框架，通过小波分析增强非语言信息，提升多模态意图识别精度。 multimodal
4 Large Language Models Miss the Multi-Agent Mark 批判性分析：大型语言模型在多智能体系统应用中偏离理论基础 large language model
5 Complex System Diagnostics Using a Knowledge Graph-Informed and Large Language Model-Enhanced Framework 提出基于知识图谱与大语言模型增强的复杂系统诊断框架，提升核电站等高可靠性系统诊断能力。 large language model
6 Position is Power: System Prompts as a Mechanism of Bias in Large Language Models (LLMs) 揭示LLM系统提示位置偏差：人口统计信息位置影响模型决策 large language model
7 StreamLink: Large-Language-Model Driven Distributed Data Engineering System StreamLink：基于大语言模型的分布式数据工程系统，提升数据处理效率与用户体验。 large language model
8 CoderAgent: Simulating Student Behavior for Personalized Programming Learning with Large Language Models 提出CoderAgent，模拟学生编程行为，实现个性化编程学习 large language model
9 Comparisons between a Large Language Model-based Real-Time Compound Diagnostic Medical AI Interface and Physicians for Common Internal Medicine Cases using Simulated Patients 基于大型语言模型的实时复合诊断医疗AI在内科常见病例中表现优于医生 large language model
10 MME-Reasoning: A Comprehensive Benchmark for Logical Reasoning in MLLMs MME-Reasoning：一个用于评估多模态大语言模型逻辑推理能力的综合基准 large language model multimodal
11 Beyond Chemical QA: Evaluating LLM's Chemical Reasoning with Modular Chemical Operations ChemCoTBench：通过模块化化学操作评估LLM的化学推理能力 large language model chain-of-thought
12 Policy Induction: Predicting Startup Success via Explainable Memory-Augmented In-Context Learning 提出基于可解释记忆增强上下文学习的策略归纳方法，预测初创公司成功率。 large language model
13 Scientific Paper Retrieval with LLM-Guided Semantic-Based Ranking SemRank：利用LLM引导的语义排序进行科学论文检索 large language model
14 Make Planning Research Rigorous Again! 强调严谨性：将传统规划的经验融入大语言模型规划，避免重复错误。 large language model
15 The Feasibility of Topic-Based Watermarking on Academic Peer Reviews 提出基于主题的水印方法，用于学术同行评议中LLM生成文本的溯源。 large language model
16 The Multilingual Divide and Its Impact on Global AI Safety 揭示多语言AI能力差距，强调其对全球AI安全的影响与挑战 large language model
17 Breaking the Ceiling: Exploring the Potential of Jailbreak Attacks through Expanding Strategy Space 通过扩展策略空间突破大型语言模型越狱攻击的性能上限 large language model ✅
18 Interpreting Social Bias in LVLMs via Information Flow Analysis and Multi-Round Dialogue Evaluation 提出信息流分析与多轮对话评估框架，用于解释LVLMs中的社会偏见。 multimodal
19 Herd Behavior: Investigating Peer Influence in LLM-based Multi-Agent Systems 研究LLM多智能体系统中群体行为，揭示同伴影响机制并实现可控协作。 large language model
20 Agent-Environment Alignment via Automated Interface Generation 提出ALIGN框架，通过自动生成接口缓解LLM Agent与环境的错位问题 large language model ✅
21 AITEE -- Agentic Tutor for Electrical Engineering AITEE：面向电气工程的Agentic Tutor，提升个性化学习与领域知识应用 large language model
22 Towards Conversational Development Environments: Using Theory-of-Mind and Multi-Agent Architectures for Requirements Refinement 提出AlignMind，利用心智理论和多智能体架构改进软件需求精化 foundation model
23 RepoMaster: Autonomous Exploration and Understanding of GitHub Repositories for Complex Task Solving RepoMaster：自主探索和理解GitHub仓库，解决复杂任务 large language model ✅
24 Step-Wise Formal Verification for LLM-Based Mathematical Problem Solving 提出MATH-VF框架，用于形式化验证LLM数学问题求解过程的正确性。 large language model
25 Respond to Change with Constancy: Instruction-tuning with LLM for Non-I.I.D. Network Traffic Classification 提出ETooL模型，利用LLM指令调优解决非独立同分布网络流量分类难题 large language model
26 An LLM-as-Judge Metric for Bridging the Gap with Human Evaluation in SE Tasks 提出SE-Jury，一种基于LLM集成裁判的软件工程任务评估指标，更贴近人工评估。 large language model
27 Code Researcher: Deep Research Agent for Large Systems Code and Commit History 提出Code Researcher：用于大型系统代码和提交历史的深度研究Agent large language model
28 GIFARC: Synthetic Dataset for Leveraging Human-Intuitive Analogies to Elevate AI Reasoning 提出GIFARC：利用人类直觉类比提升AI推理能力的合成数据集 large language model
29 MIRROR: Multi-agent Intra- and Inter-Reflection for Optimized Reasoning in Tool Learning 提出MIRROR框架以优化工具学习中的多智能体反思机制 large language model

🔬 支柱二：RL算法与架构 (RL & Architecture) (13 篇)

#题目一句话要点标签🔗⭐
30 Large Language Model-enhanced Reinforcement Learning for Low-Altitude Economy Networking 提出LLM增强的强化学习框架，解决低空经济网络复杂决策问题 reinforcement learning reward design large language model
31 Aligning Proteins and Language: A Foundation Model for Protein Retrieval 提出一种基于对比学习的蛋白质-语言对齐框架，用于蛋白质结构的功能检索。 contrastive learning foundation model multimodal
32 RLJP: Legal Judgment Prediction via First-Order Logic Rule-enhanced with Large Language Models 提出RLJP：一种基于一阶逻辑规则增强的大语言模型法律判决预测框架 contrastive learning large language model
33 Learning optimal treatment strategies for intraoperative hypotension using deep reinforcement learning 利用深度强化学习优化术中低血压的治疗策略 reinforcement learning deep reinforcement learning
34 R1-Code-Interpreter: LLMs Reason with Code via Supervised and Multi-stage Reinforcement Learning R1-Code-Interpreter：通过监督学习和多阶段强化学习，提升LLM的代码推理能力 reinforcement learning curriculum learning large language model ✅
35 Why Distillation can Outperform Zero-RL: The Role of Flexible Reasoning 仅用少量数据，蒸馏法在LLM推理能力上超越Zero-RL reinforcement learning distillation large language model
36 Revisiting Multi-Agent World Modeling from a Diffusion-Inspired Perspective DIMA：利用扩散模型提升多智能体世界建模的性能与鲁棒性 reinforcement learning policy learning world model ✅
37 LLM-Guided Reinforcement Learning: Addressing Training Bottlenecks through Policy Modulation 提出LLM引导的强化学习策略调制框架，解决训练瓶颈问题 reinforcement learning large language model
38 Reinforcement Learning-based Sequential Route Recommendation for System-Optimal Traffic Assignment 提出基于强化学习的序贯路径推荐方法，实现系统最优交通分配 reinforcement learning deep reinforcement learning
39 Bridging the Gap: Self-Optimized Fine-Tuning for LLM-based Recommender Systems 提出自优化微调SOFT方法，弥合LLM在推荐系统中的知识鸿沟 curriculum learning distillation large language model
40 Towards Safety Reasoning in LLMs: AI-agentic Deliberation for Policy-embedded CoT Data Creation 提出AIDSAFE，通过多智能体迭代审议生成策略嵌入的CoT数据，提升LLM安全性。 DPO chain-of-thought ✅
41 RRO: LLM Agent Optimization Through Rising Reward Trajectories 提出RRO：通过提升奖励轨迹优化LLM Agent，解决复杂多步任务难题。 reinforcement learning large language model
42 A Reinforcement-Learning-Enhanced LLM Framework for Automated A/B Testing in Personalized Marketing 提出RL-LLM-ABTest框架，用于个性化营销中自动化A/B测试，提升用户响应。 reinforcement learning

🔬 支柱三：空间感知与语义 (Perception & Semantics) (2 篇)

#题目一句话要点标签🔗⭐
43 Wideband RF Radiance Field Modeling Using Frequency-embedded 3D Gaussian Splatting 提出基于频率嵌入3D高斯溅射的宽带射频辐射场建模方法，解决多频段射频信号统一建模问题。 3D gaussian splatting 3DGS gaussian splatting
44 Assured Autonomy with Neuro-Symbolic Perception 提出神经符号感知框架NeuSPaPer，提升网络物理系统在对抗环境下的可靠性。 scene understanding foundation model

🔬 支柱一：机器人控制 (Robot Control) (2 篇)

#题目一句话要点标签🔗⭐
45 Text-Queried Audio Source Separation via Hierarchical Modeling 提出HSM-TSS，通过分层建模实现文本查询的音频源分离，提升语义一致性和数据效率。 manipulation
46 ADA: Automated Moving Target Defense for AI Workloads via Ephemeral Infrastructure-Native Rotation in Kubernetes ADA：基于Kubernetes的AI工作负载自动化移动目标防御系统 manipulation

⬅️ 返回 cs.AI 首页 · 🏠 返回主页