cs.AI（2025-12-07）

📊 共 24 篇论文 | 🔗 4 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (17 🔗3) 支柱二：RL算法与架构 (RL & Architecture) (6 🔗1) 支柱一：机器人控制 (Robot Control) (1)

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

#题目一句话要点标签🔗⭐
1 Singing Timbre Popularity Assessment Based on Multimodal Large Foundation Model 提出VocalVerse，基于多模态大模型进行无参考、多维度的歌唱音色流行度评估。 large language model foundation model multimodal
2 Latency-Response Theory Model: Evaluating Large Language Models via Response Accuracy and Chain-of-Thought Length 提出延迟-响应理论模型(LaRT)，通过响应准确率和思维链长度评估大语言模型。 large language model chain-of-thought ✅
3 LoopBench: Discovering Emergent Symmetry Breaking Strategies with LLM Swarms 提出LoopBench以评估LLM在分布式对称破缺中的推理能力 large language model
4 Optimal and Diffusion Transports in Machine Learning 统一框架分析机器学习中的概率分布演化，涵盖扩散模型与最优传输 large language model
5 Reformulate, Retrieve, Localize: Agents for Repository-Level Bug Localization 提出基于LLM的智能体，通过查询重构提升代码仓库级缺陷定位 large language model
6 ELANA: A Simple Energy and Latency Analyzer for LLMs ELANA：一款轻量级的LLM能耗与延迟分析工具，支持多GPU和边缘设备。 large language model ✅
7 Permission Manifests for Web Agents 提出agent-permissions.json，解决LLM驱动Web Agent权限管理难题。 large language model
8 SoK: Trust-Authorization Mismatch in LLM Agent Interactions 构建LLM Agent交互安全框架，揭示信任-授权不匹配问题 large language model
9 BabelCoder: Agentic Code Translation with Specification Alignment BabelCoder：提出基于Agent协作的代码翻译框架，提升代码迁移的准确性。 large language model
10 Do Persona-Infused LLMs Affect Performance in a Strategic Reasoning Game? 研究人格化LLM在战略推理游戏中的表现，发现特定人格能提升决策能力。 large language model
11 Formal that "Floats" High: Formal Verification of Floating Point Arithmetic 提出一种可扩展的RTL级浮点算术形式化验证方法，结合AI辅助提升验证效率。 large language model
12 Leveraging LLMs to support co-evolution between definitions and instances of textual DSLs 利用大型语言模型支持文本DSL定义与实例的协同演化 large language model
13 From Description to Score: Can LLMs Quantify Vulnerabilities? 利用大型语言模型量化漏洞：从描述到CVSS评分的自动化 large language model
14 DoVer: Intervention-Driven Auto Debugging for LLM Multi-Agent Systems DoVer：基于干预的LLM多智能体系统自动调试框架 large language model
15 ProAgent: Harnessing On-Demand Sensory Contexts for Proactive LLM Agent Systems ProAgent：利用按需感知上下文实现主动式LLM Agent系统 large language model
16 Cognitive Control Architecture (CCA): A Lifecycle Supervision Framework for Robustly Aligned AI Agents 提出认知控制架构CCA，解决LLM Agent中IPI攻击的鲁棒对齐问题 large language model
17 Stochasticity in Agentic Evaluations: Quantifying Inconsistency with Intraclass Correlation 提出使用类内相关系数(ICC)量化Agent评估中的随机性，提升评估可靠性。 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 JT-DA: Enhancing Data Analysis with Tool-Integrated Table Reasoning Large Language Models JT-DA：通过工具集成表格推理大语言模型增强数据分析能力 reinforcement learning large language model foundation model
19 Decouple to Generalize: Context-First Self-Evolving Learning for Data-Scarce Vision-Language Reasoning 提出DoGe框架，通过解耦学习上下文和问题求解，提升数据稀缺场景下VLM的泛化能力 reinforcement learning curriculum learning multimodal
20 On Memory: A comparison of memory mechanisms in world models 研究Transformer世界模型中的记忆机制，提升长时规划能力 world model
21 Predictive Modeling of I/O Performance for Machine Learning Training Pipelines: A Data-Driven Approach to Storage Optimization 提出基于机器学习的I/O性能预测模型，优化机器学习训练pipeline的存储配置。 predictive model ✅
22 Towards Small Language Models for Security Query Generation in SOC Workflows 提出面向安全运营中心工作流的小型语言模型KQL查询生成方法，降低查询成本。 distillation chain-of-thought
23 LightSearcher: Efficient DeepSearch via Experiential Memory LightSearcher：通过经验记忆实现高效的深度搜索 reinforcement learning reward shaping

🔬 支柱一：机器人控制 (Robot Control) (1 篇)

#题目一句话要点标签🔗⭐
24 PrivLLMSwarm: Privacy-Preserving LLM-Driven UAV Swarms for Secure IoT Surveillance PrivLLMSwarm：面向安全物联网监控的隐私保护LLM驱动无人机群 MPC reinforcement learning large language model

⬅️ 返回 cs.AI 首页 · 🏠 返回主页