cs.AI（2026-11-03）

📊 共 39 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) (21 🔗2) 支柱二：RL算法与架构 (RL & Architecture) (11 🔗2) 支柱一：机器人控制 (Robot Control) (2) 支柱七：动作重定向 (Motion Retargeting) (1) 支柱八：物理动画 (Physics-based Animation) (1) 支柱五：交互与反应 (Interaction & Reaction) (1) 支柱三：空间感知与语义 (Perception & Semantics) (1) 支柱六：视频提取与匹配 (Video Extraction) (1)

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

#题目一句话要点标签🔗⭐
1 GlucoAgent: Personalized Glucose Forecasting from Multimodal Personal Context 提出GlucoAgent以解决个性化血糖预测问题 foundation model multimodal
2 Feeling of Knowing in Large Language Models 提出功能性知识感知通道以解决大语言模型的知识判断不一致问题 large language model
3 The Platonic Brain: Representational Convergence in EEG Foundation Models 提出脑电图基础模型以验证普拉托尼克表征假设 foundation model
4 DGC-Net: Cross-Modal Dynamic Routing with Global-Guided Compensation in Multimodal Sentiment Analysis 提出DGC-Net以解决多模态情感分析中的跨模态不一致问题 multimodal
5 Large Language Model-Assisted Adaptive Heuristic Search for Agile Earth Observation Satellite Scheduling 提出大语言模型辅助的自适应启发式搜索以解决敏捷地球观测卫星调度问题 large language model
6 AceAction-5M: Hierarchical Goals for Generalist Action Learning Across 3D Game Worlds 提出AceAction数据集以解决3D游戏中的行动学习问题 vision-language-action VLA
7 Adversarial Speculative Reasoning Improves Long Horizon Agents 提出对抗性推测推理以提升长时间跨度智能体的推理能力 large language model
8 The Geometry of Ignorance: Detecting Suppressed Knowledge in LLMs through Activation Fragmentation 提出几何一致性框架以检测LLM中的知识抑制问题 large language model ✅
9 Creating Power Flexibility in Datacenters via Query Routing 通过查询路由实现数据中心的电力灵活性 large language model
10 From Opinions to Commitments: Unstable Assignments and Silent Choices in AI-Mediated Proposals 提出审计机制以解决AI中提案的选择不稳定性问题 large language model
11 LatentTriad: Pre-Action Risk Representations for Malicious Agent Skill Detection 提出LatentTriad以解决恶意代理技能检测中的风险问题 large language model
12 ALBI: LLM-Guided Inductive Logic Programming via Iterative Generate-Test-Constrain Refineme 提出ALBI框架以提升归纳逻辑编程的效率与准确性 large language model
13 The Pen is Mightier than the Pixel: Natural Textual Unlearnable Example for Large Vision Language Models 提出自然文本不可学习示例以保护大规模视觉语言模型 multimodal
14 Asymptotic Optimality of $cμ$-Rule for LLM Inference Scheduling under Memory Constraint 提出$cμ$规则以解决LLM推理调度中的内存约束问题 large language model
15 Are LLM Assistants Regulatory Ready? Answers from regenold's 2026 EU AI Act Q&A Challenge 提出EU AI Act Q&A基准以评估LLM助手的合规性 large language model ✅
16 From Watching to Steering: Enabling Human-Coding Agent Collaboration through Execution Traces 提出Vibetrace以解决人机协作中的代码执行可视化问题 large language model
17 FP4Quant: Efficient 4-Bit Communication Compression for Model-Parallel LLM Training 提出FP4Quant以解决大规模语言模型训练中的通信瓶颈问题 large language model
18 Planning-Coding Separation in LLM-Based Code Generation: An Empirical Study 提出规划与编码分离以提升LLM代码生成效果 large language model
19 Reliability Stress Testing of RAG Retrieval under Query-Relevant Corpus Poisoning 提出RAG检索系统的可靠性压力测试方法以应对查询相关的语料中毒问题 instruction following
20 When the Consumer Is an Agent: Execution-Grounded Service-Level Agreements 提出执行基础的完成合同以解决异步服务中的任务完成问题 large language model
21 Validation-Gated ML Pipeline Synthesis with Expert-Grounded Self-Evolving Skills 提出EvoML以解决现有AutoML方法的适应性和可靠性问题 large language model

🔬 支柱二：RL算法与架构 (RL & Architecture) (11 篇)

#题目一句话要点标签🔗⭐
22 Embodied AI for Scientific Discovery: From Physical Action to Scientific Evidence 提出一种体现AI的科学发现方法以解决实验选择与执行问题 world model world models embodied AI
23 A research world model of power engineering learns its dynamics and reveals what-ifs 提出研究世界模型以模拟电力工程研究决策的动态 world model world models latent dynamics
24 Reconstructing Fraudulent Strategies in Online Payments via Interpretable Inverse Reinforcement Learning 提出PaySeqGen框架以重建在线支付中的欺诈策略 reinforcement learning inverse reinforcement learning
25 Think Less, Locally: Local Preference Learning for Reasoning Compression 提出LoCo以解决推理模型的过度思考问题 preference learning DPO chain-of-thought ✅
26 JHiMA-WebRL: A Joint Hierarchical Multi-Agent Reinforcement Learning Framework for Long-Horizon Web Tasks 提出JHiMA-WebRL框架以解决长时间网页任务的规划挑战 reinforcement learning multimodal
27 Catalyzing Interactive HTML Generation Capabilities via High-Fidelity Reinforcement Learning 提出WebCatalyst框架以解决交互式HTML生成中的功能与视觉质量问题 reinforcement learning large language model
28 UniFlow: Unified Multi-Scale Flow Matching for Time Series Generation 提出UniFlow以解决多尺度时间序列生成问题 flow matching representation learning ✅
29 AgentAM: An AI Agent for Spatter- and Defect-Aware Process Optimization in Laser Powder Bed Fusion 提出AgentAM以解决激光粉末床熔融过程优化问题 predictive model large language model
30 CogniLife: Continuous Cognitive Distillation for Deployed Multi-Expert Agent Marketplaces 提出CogniLife以解决多专家代理市场中的冷专家问题 distillation
31 Bidirectional Centered Score Matching for Discrete Diffusion Distillation 提出双向中心评分匹配以优化离散扩散蒸馏 distillation
32 CORA: Teacher-Guided Causal O-RAN Adaptation for Jamming Detection 提出CORA框架以解决无标签干扰检测问题 reinforcement learning PPO

🔬 支柱一：机器人控制 (Robot Control) (2 篇)

#题目一句话要点标签🔗⭐
33 Verifier strength does not buy discovery quality: a pre-registered two-domain test of LLM-based automated research, with post-hoc evidence pointing to proposer-pool composition 提出基于LLM的自动化研究方法以优化发现质量 manipulation
34 When Debate Amplifies Risk: Social-Bias Vulnerabilities in Multi-Agent Safety Deliberation 提出RiskTrace-Debias以解决多代理辩论中的安全风险问题 manipulation

🔬 支柱七：动作重定向 (Motion Retargeting) (1 篇)

#题目一句话要点标签🔗⭐
35 Human-to-Robot Transfer via Bridging Representation and Action Space in VLAs 提出H2RB框架以解决人机示范适应问题 human-to-robot vision-language-action VLA

🔬 支柱八：物理动画 (Physics-based Animation) (1 篇)

#题目一句话要点标签🔗⭐
36 Learning Behaviorally Relevant Dynamics in Multimodal Neural Recordings 提出BREM-NET以解决多模态神经数据分析问题 spatiotemporal multimodal

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

#题目一句话要点标签🔗⭐
37 Perseus: A Full-Stack System for Homomorphic Model Inference 提出Perseus以实现高效同态模型推理 OMOMO large language model

🔬 支柱三：空间感知与语义 (Perception & Semantics) (1 篇)

#题目一句话要点标签🔗⭐
38 DualWorldBench: Can Agents Plan Deliveries across Symbolic and Grounded Worlds? 提出DualWorldBench以解决多目标规划中的感知与推理整合问题 scene understanding multimodal

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

#题目一句话要点标签🔗⭐
39 Spatial Memory for AR Glasses: Hierarchical Visual Acquisition and Source-Linked Recall 提出空间记忆系统以增强AR眼镜的历史回忆能力 egocentric

⬅️ 返回 cs.AI 首页 · 🏠 返回主页