cs.AI（2026-01-19）

📊 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) (18 🔗3) 支柱二：RL算法与架构 (RL & Architecture) (4 🔗1) 支柱六：视频提取与匹配 (Video Extraction) (1) 支柱一：机器人控制 (Robot Control) (1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (18 篇)

#题目一句话要点标签🔗⭐
1 The Geometry of Thought: How Scale Restructures Reasoning In Large Language Models 揭示大语言模型推理的几何结构：规模扩展重塑推理方式 large language model chain-of-thought
2 KOCO-BENCH: Can Large Language Models Leverage Domain Knowledge in Software Development? KOCO-BENCH：评估大语言模型在领域知识驱动的软件开发中的能力 large language model ✅
3 Explicit Cognitive Allocation: A Principle for Governed and Auditable Inference in Large Language Models 提出显式认知分配原则，提升大语言模型推理过程的可控性和可追溯性 large language model
4 Integrating Virtual Reality and Large Language Models for Team-Based Non-Technical Skills Training and Evaluation in the Operating Room VORTeX：结合VR与LLM，用于手术室团队非技术技能培训与评估 large language model
5 A Lightweight Modular Framework for Constructing Autonomous Agents Driven by Large Language Models: Design, Implementation, and Applications in AgentForge AgentForge：轻量级模块化框架，赋能大语言模型驱动的自主Agent构建 large language model
6 Scientific production in the era of Large Language Models 大型语言模型显著提升科研论文产量，但可能降低论文质量并改变引用模式 large language model
7 CORVUS: Red-Teaming Hallucination Detectors via Internal Signal Camouflage in Large Language Models CORVUS：通过内部信号伪装对抗大语言模型幻觉检测器 large language model
8 An Evolutionary Framework for Automatic Optimization Benchmark Generation via Large Language Models 提出进化框架以自动生成优化基准测试 large language model
9 Vision Language Models for Optimization-Driven Intent Processing in Autonomous Networks IntentOpt：评估视觉语言模型在自治网络中优化驱动意图处理的能力 large language model multimodal
10 Tracing the Data Trail: A Survey of Data Provenance, Transparency and Traceability in LLMs 综述LLM数据溯源、透明性和可追溯性，填补训练数据生命周期不透明的空白。 large language model
11 SCULPT: Constraint-Guided Pruned MCTS that Carves Efficient Paths for Mathematical Reasoning SCULPT：约束引导的剪枝MCTS，为数学推理规划高效路径 large language model
12 Real-Time Deadlines Reveal Temporal Awareness Failures in LLM Strategic Dialogues 揭示LLM在战略对话中对实时截止时间的感知缺陷 large language model
13 Prompt Injection Mitigation with Agentic AI, Nested Learning, and AI Sustainability via Semantic Caching 提出基于Agentic AI、嵌套学习和语义缓存的提示注入缓解方法，提升LLM安全性与可持续性。 large language model
14 ArchAgent: Scalable Legacy Software Architecture Recovery with LLMs ArchAgent：利用LLM实现大规模遗留软件架构的可扩展恢复 large language model ✅
15 Beyond Accuracy: Characterizing Code Comprehension Capabilities in (Large) Language Models 提出诊断框架以评估大型语言模型的代码理解能力 large language model
16 On the Evidentiary Limits of Membership Inference for Copyright Auditing 研究表明，针对LLM的成员推断攻击在版权审计中证据力不足 large language model
17 MirrorGuard: Toward Secure Computer-Use Agents via Simulation-to-Real Reasoning Correction MirrorGuard：通过模拟到真实推理校正增强计算机使用代理的安全性 foundation model ✅
18 Empowering All-in-Loop Health Management of Spacecraft Power System in the Mega-Constellation Era via Human-AI Collaboration 提出SpaceHMchat人机协作框架，赋能巨型星座时代航天器电源系统全环健康管理 TAMP

🔬 支柱二：RL算法与架构 (RL & Architecture) (4 篇)

#题目一句话要点标签🔗⭐
19 STEP-LLM: Generating CAD STEP Models from Natural Language with Large Language Models STEP-LLM：利用大语言模型从自然语言生成CAD STEP模型，提升几何保真度。 reinforcement learning large language model chain-of-thought
20 CURE-Med: Curriculum-Informed Reinforcement Learning for Multilingual Medical Reasoning 提出CURE-MED框架，解决LLM在多语言医疗推理中逻辑性和语言一致性问题。 reinforcement learning large language model
21 Communication Methods in Multi-Agent Reinforcement Learning 综述多智能体强化学习中的通信方法，分析优劣并指出未来研究方向 reinforcement learning
22 Teaching LLMs to Learn Tool Trialing and Execution through Environment Interaction ToolMaster：通过环境交互教LLM学习工具试错与执行 reinforcement learning large language model ✅

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

#题目一句话要点标签🔗⭐
23 VIRO: Robust and Efficient Neuro-Symbolic Reasoning with Verification for Referring Expression Comprehension 提出VIRO框架，通过验证机制增强神经符号推理在指代表达理解中的鲁棒性与效率。 egocentric spatial relationship large language model

🔬 支柱一：机器人控制 (Robot Control) (1 篇)

#题目一句话要点标签🔗⭐
24 Neurosymbolic LoRA: Why and When to Tune Weights vs. Rewrite Prompts 提出神经符号LoRA框架，动态融合数值微调与符号编辑以提升LLM适应性 manipulation large language model

⬅️ 返回 cs.AI 首页 · 🏠 返回主页