cs.AI（2025-07-14）

📊 共 34 篇论文 | 🔗 2 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (21 🔗2) 支柱二：RL算法与架构 (RL & Architecture) (10) 支柱一：机器人控制 (Robot Control) (1) 支柱五：交互与反应 (Interaction & Reaction) (1) 支柱四：生成式动作 (Generative Motion) (1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (21 篇)

#题目一句话要点标签🔗⭐
1 Deep Hidden Cognition Facilitates Reliable Chain-of-Thought Reasoning 提出基于深度隐藏认知的CoT推理校准方法，提升推理可靠性 large language model multimodal chain-of-thought
2 React to This (RTT): A Nonverbal Turing Test for Embodied AI 提出React to This (RTT)非语言图灵测试，评估具身AI的交互感知和可信度 embodied AI
3 Enhancing the Capabilities of Large Language Models for API calls through Knowledge Graphs 提出KG2data，利用知识图谱增强大语言模型在气象领域API调用能力 large language model
4 Chat with AI: The Surprising Turn of Real-time Video Communication from Human to AI 针对AI视频聊天，提出上下文感知视频流以降低延迟并保持MLLM准确性 large language model multimodal ✅
5 Automating SPARQL Query Translations between DBpedia and Wikidata 利用大型语言模型自动翻译DBpedia和Wikidata之间的SPARQL查询 large language model chain-of-thought
6 How Robust are LLM-Generated Library Imports? An Empirical Study using Stack Overflow 评估LLM生成代码中库导入的鲁棒性：基于Stack Overflow的实证研究 large language model
7 Logic layer Prompt Control Injection (LPCI): A Novel Security Vulnerability Class in Agentic Systems 提出逻辑层提示控制注入（LPCI）攻击，揭示Agentic系统中新型安全漏洞 large language model
8 Kodezi Chronos: A Debugging-First Language Model for Repository-Scale Code Understanding Kodezi Chronos-1：面向仓库级代码理解的调试优先语言模型 large language model
9 Automated Thematic Analyses Using LLMs: Xylazine Wound Management Social Media Chatter Use Case 利用LLM自动进行主题分析：以赛拉嗪伤口管理社交媒体讨论为例 large language model
10 AI-Powered Math Tutoring: Platform for Personalized and Adaptive Education 提出基于多智能体AI的数学辅导平台，实现个性化和自适应教育 large language model
11 CodeAssistBench (CAB): Dataset & Benchmarking for Multi-turn Chat-Based Code Assistance 提出CodeAssistBench，用于评估多轮对话代码辅助的基准测试。 large language model ✅
12 From Semantic Web and MAS to Agentic AI: A Unified Narrative of the Web of Agents 构建Agentic AI：提出Web of Agents演进的统一叙事框架，弥合多智能体系统与语义网的鸿沟。 large language model
13 Evaluating LLM-based Workflows for Switched-Mode Power Supply Design 提出基于LLM的工作流，用于开关电源设计的自动化与优化 large language model
14 Natural Language-based Assessment of L2 Oral Proficiency using LLMs 利用大型语言模型进行基于自然语言的二语口语能力评估 large language model
15 Breaking the Myth: Can Small Models Infer Postconditions Too? 小模型亦可胜任：微调7B模型实现高质量后置条件推断 large language model
16 Differentially Private Federated Low Rank Adaptation Beyond Fixed-Matrix 提出FedASK，解决差分隐私联邦LoRA中适配器更新与隐私保护的难题 large language model
17 Modeling the Economic Impacts of AI Openness Regulation 构建AI开放性监管的经济影响模型，评估不同开放标准对开发者激励的影响 foundation model
18 DeepSeek: Paradigm Shifts and Technical Evolution in Large AI Models DeepSeek发布V3和R1系列大模型，通过创新算法和工程优化实现低成本高性能。 large language model
19 Turning the Tide: Repository-based Code Reflection 提出LiveRepoReflection基准与RepoReflection-Instruct数据集，提升代码大模型在代码仓库中的代码反思能力 large language model
20 Model-Grounded Symbolic Artificial Intelligence Systems Learning and Reasoning with Model-Grounded Symbolic Artificial Intelligence Systems 将指令调优大语言模型重释为模型基底符号AI系统，提升学习效率和推理可靠性。 large language model
21 Game Theory Meets LLM and Agentic AI: Reimagining Cybersecurity for the Age of Intelligent Threats 博弈论结合LLM与Agentic AI，重塑智能威胁时代的网络安全 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🔗⭐
22 Multimodal Fusion at Three Tiers: Physics-Driven Data Generation and Vision-Language Guidance for Brain Tumor Segmentation 提出三层融合架构，结合物理建模数据生成与视觉-语言引导，提升脑肿瘤分割精度。 contrastive learning distillation spatial relationship
23 A Code Comprehension Benchmark for Large Language Models for Code 提出代码理解基准，通过微调提升大语言模型在代码语义理解任务上的性能 DPO large language model
24 The Challenge of Teaching Reasoning to LLMs Without RL or Distillation 仅用少量高质量CoT样本，即可显著提升基座模型的推理能力 reinforcement learning distillation chain-of-thought
25 FRSICL: LLM-Enabled In-Context Learning Flight Resource Allocation for Fresh Data Collection in UAV-Assisted Wildfire Monitoring 提出FRSICL，利用LLM上下文学习解决无人机野火监测中飞行资源分配问题 reinforcement learning deep reinforcement learning DRL
26 Adaptability in Multi-Agent Reinforcement Learning: A Framework and Unified Review 提出MARL适应性评估框架，提升多智能体系统在动态环境中的可靠性。 reinforcement learning
27 PRM-Free Security Alignment of Large Models via Red Teaming and Adversarial Training 提出一种无需过程奖励模型的红队对抗训练框架，提升大模型的安全性。 curriculum learning large language model
28 Improving monotonic optimization in heterogeneous multi-agent reinforcement learning with optimal marginal deterministic policy gradient 提出OMDPG算法，解决异构多智能体强化学习中单调提升与参数共享的冲突问题。 reinforcement learning
29 UWB Radar-based Heart Rate Monitoring: A Transfer Learning Approach 提出基于迁移学习的UWB雷达心率监测方法，加速消费电子集成 MAE PULSE
30 Aligning Generative Speech Enhancement with Perceptual Feedback 提出基于感知反馈对齐的生成式语音增强方法，提升语音质量。 DPO direct preference optimization
31 AudioMAE++: learning better masked audio representations with SwiGLU FFNs AudioMAE++：利用SwiGLU FFNs学习更优的音频掩码自编码表示 masked autoencoder MAE

🔬 支柱一：机器人控制 (Robot Control) (1 篇)

#题目一句话要点标签🔗⭐
32 Should We Ever Prefer Decision Transformer for Offline Reinforcement Learning? 针对稀疏奖励离线强化学习，过滤行为克隆方法优于决策Transformer locomotion manipulation reinforcement learning

🔬 支柱五：交互与反应 (Interaction & Reaction) (1 篇)

#题目一句话要点标签🔗⭐
33 Secure and Efficient UAV-Based Face Detection via Homomorphic Encryption and Edge Computing 提出基于同态加密和边缘计算的安全高效无人机人脸检测方案 OMOMO

🔬 支柱四：生成式动作 (Generative Motion) (1 篇)

#题目一句话要点标签🔗⭐
34 AI and the Net-Zero Journey: Energy Demand, Emissions, and the Potential for Transition 分析AI发展对能源需求和碳排放的影响，探讨其在净零排放中的潜力 penetration

⬅️ 返回 cs.AI 首页 · 🏠 返回主页