cs.AI（2025-07-11）

📊 共 23 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) (17 🔗2) 支柱二：RL算法与架构 (RL & Architecture) (6 🔗1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (17 篇)

#题目一句话要点标签🔗⭐
1 From Classical Machine Learning to Emerging Foundation Models: Review on Multimodal Data Integration for Cancer Research 综述多模态数据融合在癌症研究中的应用：从传统机器学习到新兴的Foundation Models foundation model multimodal
2 BioAnalyst: A Foundation Model for Biodiversity BioAnalyst：首个面向欧洲生物多样性分析的多模态基础模型 foundation model multimodal
3 Can Large Language Models Understand As Well As Apply Patent Regulations to Pass a Hands-On Patent Attorney Test? 评估大型语言模型在专利律师考试中的表现，揭示其理解与应用专利法规的局限性 large language model multimodal
4 Hybrid Systolic Array Accelerator with Optimized Dataflow for Edge Large Language Model Inference 提出混合 systolic 阵列加速器，优化边缘大语言模型推理的数据流 large language model
5 Bridging Literature and the Universe Via A Multi-Agent Large Language Model System SimAgents：利用多Agent LLM系统桥接文献与宇宙学模拟，加速科研。 large language model ✅
6 AMix-1: A Pathway to Test-Time Scalable Protein Foundation Model AMix-1：一种可测试时扩展的蛋白质基础模型，提升蛋白质工程能力 foundation model
7 Introspection of Thought Helps AI Agents 提出INoT框架，通过LLM内部自省推理降低AI Agent的token成本并提升性能 large language model multimodal chain-of-thought
8 Agentic Large Language Models for Conceptual Systems Engineering and Design 提出基于Agentic LLM的概念系统工程设计方法，提升设计细节粒度。 large language model
9 Invariant-based Robust Weights Watermark for Large Language Models 提出一种基于不变性的鲁棒权重水印方案，用于保护大语言模型的知识产权。 large language model
10 Large Multi-modal Model Cartographic Map Comprehension for Textual Locality Georeferencing 提出一种基于大模型多模态理解的地图制图方法，用于地名文本地理定位。 large language model
11 SetupBench: Assessing Software Engineering Agents' Ability to Bootstrap Development Environments SetupBench：评估软件工程Agent引导开发环境的能力 large language model
12 ToolRegistry: A Protocol-Agnostic Tool Management Library for Function-Calling LLMs ToolRegistry：为函数调用LLM提供协议无关的工具管理库 large language model ✅
13 How to Train a Leader: Hierarchical Reasoning in Multi-Agent LLMs 提出MLPO，训练单个领导者LLM协调多智能体推理，提升复杂任务性能 large language model
14 GraphRunner: A Multi-Stage Framework for Efficient and Accurate Graph-Based Retrieval GraphRunner：一种高效准确的图检索多阶段框架，解决LLM推理错误和幻觉问题。 large language model
15 Optimizing Sequential Multi-Step Tasks with Parallel LLM Agents M1-Parallel：并行LLM智能体加速多步骤序列任务 large language model
16 ARPaCCino: An Agentic-RAG for Policy as Code Compliance 提出ARPaCCino，利用Agentic-RAG实现策略即代码的合规性自动化验证与生成。 large language model
17 Findings of the BEA 2025 Shared Task on Pedagogical Ability Assessment of AI-powered Tutors BEA 2025共享任务：评估AI辅导系统在教育对话中教学能力 large language model

🔬 支柱二：RL算法与架构 (RL & Architecture) (6 篇)

#题目一句话要点标签🔗⭐
18 Safe Deep Reinforcement Learning for Resource Allocation with Peak Age of Information Violation Guarantees 提出基于优化理论的安全深度强化学习框架，解决无线网络控制系统资源分配问题。 reinforcement learning deep reinforcement learning DRL
19 Leanabell-Prover-V2: Verifier-integrated Reasoning for Formal Theorem Proving via Reinforcement Learning Leanabell-Prover-V2：通过强化学习和验证器集成推理，提升形式化定理证明能力 reinforcement learning large language model chain-of-thought ✅
20 Agent Safety Alignment via Reinforcement Learning 提出基于强化学习的Agent安全对齐框架，解决工具使用Agent的安全风险。 reinforcement learning reward shaping large language model
21 M2-Reasoning: Empowering MLLMs with Unified General and Spatial Reasoning M2-Reasoning：通过统一的通用和空间推理增强多模态大语言模型 reinforcement learning large language model multimodal
22 elsciRL: Integrating Language Solutions into Reinforcement Learning Problem Settings elsciRL：集成语言解决方案到强化学习问题设置的开源库 reinforcement learning
23 White-Basilisk: A Hybrid Model for Code Vulnerability Detection White-Basilisk：一种用于代码漏洞检测的混合模型，参数量仅200M，性能超越大型语言模型。 Mamba large language model

⬅️ 返回 cs.AI 首页 · 🏠 返回主页