cs.AI（2026-04-07）

📊 共 130 篇论文 | 🔗 5 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (89 🔗4) 支柱二：RL算法与架构 (RL & Architecture) (28 🔗1) 支柱一：机器人控制 (Robot Control) (4) 支柱八：物理动画 (Physics-based Animation) (3) 支柱四：生成式动作 (Generative Motion) (2) 支柱五：交互与反应 (Interaction & Reaction) (1) 支柱七：动作重定向 (Motion Retargeting) (1) 支柱六：视频提取与匹配 (Video Extraction) (1) 支柱三：空间感知与语义 (Perception & Semantics) (1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (89 篇)

#题目一句话要点标签🔗⭐
1 A Multimodal Foundation Model of Spatial Transcriptomics and Histology for Biological Discovery and Clinical Prediction 提出STORM，一个用于生物发现和临床预测的空间转录组学和组织学多模态基础模型 foundation model multimodal
2 FeynmanBench: Benchmarking Multimodal LLMs on Diagrammatic Physics Reasoning FeynmanBench：用于评估多模态LLM在费曼图推理能力上的基准测试 large language model multimodal
3 Don't Blink: Evidence Collapse during Multimodal Reasoning 揭示多模态推理中证据崩塌现象，提出任务感知视觉否决策略 multimodal visual grounding
4 Large Language Models Align with the Human Brain during Creative Thinking 探讨大语言模型与人脑在创造性思维中的一致性 large language model chain-of-thought
5 Can LLMs Reason About Attention? Towards Zero-Shot Analysis of Multimodal Classroom Behavior 提出基于LLM的零样本多模态课堂行为分析框架，无需存储原始视频。 multimodal
6 BAAI Cardiac Agent: An intelligent multimodal agent for automated reasoning and diagnosis of cardiovascular diseases from cardiac magnetic resonance imaging BAAI Cardiac Agent：用于心血管疾病自动推理与诊断的多模态智能体 multimodal
7 Cite Pretrain: Retrieval-Free Knowledge Attribution for Large Language Models Cite Pretrain：无需检索的大语言模型知识归属方法 large language model
8 Toward Full Autonomous Laboratory Instrumentation Control with Large Language Models 利用大语言模型实现实验室仪器全自动控制 large language model
9 TABQAWORLD: Optimizing Multimodal Reasoning for Multi-Turn Table Question Answering TABQAWORLD：优化多模态推理，提升多轮表格问答性能 multimodal
10 Structured Multi-Criteria Evaluation of Large Language Models with Fuzzy Analytic Hierarchy Process and DualJudge 提出基于模糊层次分析法和DualJudge的LLM结构化多标准评估方法 large language model
11 PolySwarm: A Multi-Agent Large Language Model Framework for Prediction Market Trading and Latency Arbitrage PolySwarm：用于预测市场交易和延迟套利的LLM多智能体框架 large language model
12 The Topology of Multimodal Fusion: Why Current Architectures Fail at Creative Cognition 揭示多模态融合架构的拓扑局限性，提出基于神经ODE的拓扑正则化方法以提升创造性认知能力。 multimodal
13 CREBench: Evaluating Large Language Models in Cryptographic Binary Reverse Engineering CREBench：评估大型语言模型在密码二进制逆向工程中的能力 large language model
14 AutoReSpec: A Framework for Generating Specification using Large Language Models AutoReSpec：利用大语言模型自动生成可验证规约的协同框架 large language model
15 Enhancing behavioral nudges with large language model-based iterative personalization: A field experiment on electricity and hot-water conservation 利用大语言模型提升行为干预的个性化效果 large language model
16 Strengthening Human-Centric Chain-of-Thought Reasoning Integrity in LLMs via a Structured Prompt Framework 提出结构化提示框架，增强LLM在安全分析中类人CoT推理的完整性 chain-of-thought
17 Large Language Models for Combinatorial Optimization of Design Structure Matrix 提出基于大语言模型的DSM重排序优化方法，提升复杂工程系统模块化效率 large language model
18 TableVision: A Large-Scale Benchmark for Spatially Grounded Reasoning over Complex Hierarchical Tables TableVision：一个大规模表格基准，用于复杂分层表格上的空间推理。 large language model multimodal
19 InsTraj: Instructing Diffusion Models with Travel Intentions to Generate Real-world Trajectories 提出InsTraj以解决GPS轨迹生成的语义理解与约束问题 large language model multimodal
20 MolDA: Molecular Understanding and Generation via Large Language Diffusion Model MolDA：提出基于扩散语言模型的新型分子理解与生成框架，解决自回归模型的局限性。 large language model multimodal
21 From Paper to Program: A Multi-Stage LLM-Assisted Workflow for Accelerating Quantum Many-Body Algorithm Development 提出多阶段LLM辅助工作流以加速量子多体算法开发 large language model foundation model
22 Toward Epistemic Stability: Engineering Consistent Procedures for Industrial LLM Hallucination Reduction 提出五种Prompt工程策略，提升工业场景LLM输出的稳定性和可靠性，减少幻觉。 large language model foundation model
23 Beyond Predefined Schemas: TRACE-KG for Context-Enriched Knowledge Graphs from Complex Documents 提出TRACE-KG以解决知识图谱构建中的模式依赖问题 multimodal
24 Combee: Scaling Prompt Learning for Self-Improving Language Model Agents Combee：扩展Prompt学习，实现自提升语言模型Agent large language model
25 Soft Tournament Equilibrium 提出软锦标赛均衡(STE)框架，用于解决通用智能体评估中的循环依赖问题。 large language model
26 ShadowNPU: System and Algorithm Co-design for NPU-Centric On-Device LLM Inference ShadowNPU：面向NPU的片上LLM推理系统与算法协同设计 large language model
27 A Firefly Algorithm for Mixed-Variable Optimization Based on Hybrid Distance Modeling 提出基于混合距离建模的萤火虫算法(FAmv)以解决混合变量优化问题 multimodal
28 Evaluating Artificial Intelligence Through a Christian Understanding of Human Flourishing 提出基于基督教人类繁荣理解的AI评估框架 large language model
29 Hume's Representational Conditions for Causal Judgment: What Bayesian Formalization Abstracted Away 分析休谟因果判断理论，揭示贝叶斯形式化忽略的表征条件 large language model
30 Automated Analysis of Global AI Safety Initiatives: A Taxonomy-Driven LLM Approach 提出基于活动分类法的LLM自动分析框架，用于评估全球AI安全倡议的政策文件。 large language model
31 When Do Hallucinations Arise? A Graph Perspective on the Evolution of Path Reuse and Path Compression 通过图视角分析LLM推理幻觉的产生机制：路径复用与路径压缩 large language model
32 Single-agent vs. Multi-agents for Automated Video Analysis of On-Screen Collaborative Learning Behaviors 提出基于VLM的多智能体系统，用于自动化分析屏幕协作学习行为 multimodal
33 Beyond Retrieval: Modeling Confidence Decay and Deterministic Agentic Platforms in Generative Engine Optimization 提出确定性Agent平台，解决生成引擎优化中RAG的幻觉和零点击问题 large language model
34 Affording Process Auditability with QualAnalyzer: An Atomistic LLM Analysis Tool for Qualitative Research QualAnalyzer：通过原子化LLM分析实现定性研究过程的可审计性 large language model
35 Profile-Then-Reason: Bounded Semantic Complexity for Tool-Augmented Language Agents 提出Profile-Then-Reason框架，提升工具增强语言代理的效率与可靠性 large language model
36 Readable Minds: Emergent Theory-of-Mind-Like Behavior in LLM Poker Agents LLM扑克Agent在动态交互中涌现类心智理论行为 large language model
37 InferenceEvolve: Towards Automated Causal Effect Estimators through Self-Evolving AI InferenceEvolve：利用自进化AI实现因果效应估计器的自动发现与优化 large language model
38 AI Trust OS -- A Continuous Governance Framework for Autonomous AI Observability and Zero-Trust Compliance in Enterprise Environments 提出AI Trust OS，实现企业环境中自治AI的可观测性和零信任合规的持续治理。 large language model
39 MemMachine: A Ground-Truth-Preserving Memory System for Personalized AI Agents MemMachine：一种面向个性化AI代理的、保留真实信息的记忆系统 large language model
40 From Concept to Practice: an Automated LLM-aided UVM Machine for RTL Verification UVM^2：一种基于LLM的自动化UVM机器，用于RTL验证，显著提升验证效率。 large language model
41 The Persuasion Paradox: When LLM Explanations Fail to Improve Human-AI Team Performance 揭示LLM解释在提升人机团队表现中的悖论 large language model
42 FVRuleLearner: Operator-Level Reasoning Tree (OP-Tree)-Based Rules Learning for Formal Verification FVRuleLearner：提出基于算子推理树的规则学习框架，提升形式验证中SVA生成的正确性。 large language model
43 Scaling Teams or Scaling Time? Memory Enabled Lifelong Learning in LLM Multi-Agent Systems 提出LLMA-Mem框架，通过记忆增强提升LLM多智能体系统在长期任务中的性能和效率。 large language model
44 Toward Executable Repository-Level Code Generation via Environment Alignment 提出EnvGraph框架，通过环境对齐实现可执行的仓库级代码生成。 large language model
45 Persistent Cross-Attempt State Optimization for Repository-Level Code Generation LiveCoder：通过跨尝试状态优化提升代码仓库级代码生成效果 large language model
46 Can Humans Tell? A Dual-Axis Study of Human Perception of LLM-Generated News 研究表明人类难以区分LLM生成的新闻与人工撰写的新闻，用户侧检测防御不可行。 large language model
47 CoopGuard: Stateful Cooperative Agents Safeguarding LLMs Against Evolving Multi-Round Attacks CoopGuard：基于合作代理的状态化防御框架，抵御LLM多轮对抗攻击 large language model
48 Commercial Persuasion in AI-Mediated Conversations 研究揭示LLM驱动的对话式AI在商业推广中存在隐蔽诱导用户选择的风险 large language model
49 Similarity Field Theory: A Mathematical Framework for Intelligence 提出相似性场论，为理解智能系统提供数学框架 large language model
50 Autonomous Agents for Scientific Discovery: Orchestrating Scientists, Language, Code, and Physics 基于LLM的自主智能体加速科学发现，实现科学家、语言、代码和物理的协同。 large language model
51 Agentic AI Security: Threats, Defenses, Evaluation, and Open Challenges 针对Agentic AI的安全威胁，提出防御、评估方法与开放挑战 large language model
52 An Agent-Based Framework for the Automatic Validation of Mathematical Optimization Models 提出一种基于Agent的框架，用于自动验证数学优化模型。 large language model
53 The Paradox of Robustness: Decoupling Rule-Based Logic from Affective Noise in High-Stakes Decision-Making 大语言模型在规则约束决策中表现出对情感框架的鲁棒性，揭示了“鲁棒性悖论”。 large language model
54 IV Co-Scientist: Multi-Agent LLM Framework for Causal Instrumental Variable Discovery 提出IV Co-Scientist，利用多智能体LLM框架进行因果工具变量发现。 large language model
55 Collective AI can amplify tiny perturbations into divergent decisions 集体AI决策易受微小扰动影响，导致结果发散 large language model
56 An Onto-Relational-Sophic Framework for Governing Synthetic Minds 提出Onto-Relational-Sophic框架，用于治理通用人工智能。 foundation model
57 ATLAS: A Layered Constraint-Guided Framework for Structured Artifact Generation in LLM-Assisted MDE ATLAS：一种分层约束引导框架，用于LLM辅助的MDE中结构化工件生成 large language model
58 Beyond Message Passing: A Semantic View of Agent Communication Protocols 提出Agent通信协议的三层语义视角，揭示现有协议在语义层面的不足。 large language model
59 HybridKV: Hybrid KV Cache Compression for Efficient Multimodal Large Language Model Inference HybridKV：面向高效多模态大语言模型推理的混合KV缓存压缩框架 large language model multimodal
60 OGA-AID: Clinician-in-the-loop AI Report Drafting Assistant for Multimodal Observational Gait Analysis in Post-Stroke Rehabilitation OGA-AID：面向卒中康复的多模态步态分析临床医生辅助AI报告草拟系统 large language model multimodal
61 ETR: Entropy Trend Reward for Efficient Chain-of-Thought Reasoning 提出熵趋势奖励ETR，提升思维链推理效率与准确率 large language model chain-of-thought ✅
62 CritBench: A Framework for Evaluating Cybersecurity Capabilities of Large Language Models in IEC 61850 Digital Substation Environments 提出CritBench框架以评估IEC 61850数字变电站环境中的网络安全能力 large language model ✅
63 Context-Value-Action Architecture for Value-Driven Large Language Model Agents 提出CVA架构，通过解耦认知推理与行为生成，提升LLM Agent的价值对齐与行为可解释性。 large language model
64 Joint Knowledge Base Completion and Question Answering by Combining Large Language Models and Small Language Models 提出JCQL框架，结合大语言模型和小语言模型联合完成知识库补全与问答任务 large language model
65 JTON: A Token-Efficient JSON Superset with Zen Grid Tabular Encoding for Large Language Models 提出JTON：一种Token高效的JSON超集，采用Zen Grid表格编码，专为大型语言模型设计。 large language model
66 CAKE: Cloud Architecture Knowledge Evaluation of Large Language Models CAKE：用于评估大语言模型云架构知识的基准测试 large language model
67 QA-MoE: Towards a Continuous Reliability Spectrum with Quality-Aware Mixture of Experts for Robust Multimodal Sentiment Analysis 提出QA-MoE，通过质量感知的专家混合模型实现鲁棒的多模态情感分析 multimodal
68 From Incomplete Architecture to Quantified Risk: Multimodal LLM-Driven Security Assessment for Cyber-Physical Systems ASTRAL：利用多模态LLM进行网络物理系统架构驱动的安全风险评估 multimodal
69 From Large Language Model Predicates to Logic Tensor Networks: Neurosymbolic Offer Validation in Regulated Procurement 提出一种神经符号方法，用于在受监管的采购中验证投标文件的有效性。 large language model
70 Experience Transfer for Multimodal LLM Agents in Minecraft Game 提出Echo框架，提升多模态LLM Agent在Minecraft中经验迁移效率。 multimodal
71 Market-Bench: Benchmarking Large Language Models on Economic and Trade Competition Market-Bench：构建经济贸易竞争基准，评估大语言模型在经济活动中的能力。 large language model
72 Claw-Eval: Toward Trustworthy Evaluation of Autonomous Agents Claw-Eval：提出可信的自主Agent评估基准，解决现有评估方法的局限性。 large language model multimodal
73 LLM4CodeRE: Generative AI for Code Decompilation Analysis and Reverse Engineering LLM4CodeRE：用于代码逆向工程的双向生成式AI框架 large language model
74 How LLMs Follow Instructions: Skillful Coordination, Not a Universal Mechanism 揭示大语言模型指令遵循机制：技能协调而非通用机制 instruction following
75 Flowr -- Scaling Up Retail Supply Chain Operations Through Agentic AI in Large Scale Supermarket Chains Flowr：通过Agentic AI扩展大规模超市零售供应链运营 large language model
76 A Formal Security Framework for MCP-Based AI Agents: Threat Taxonomy, Verification Models, and Defense Mechanisms 提出MCPSHIELD框架，系统解决基于MCP的AI Agent安全威胁 large language model
77 Deep Researcher Agent: An Autonomous Framework for 24/7 Deep Learning Experimentation with Zero-Cost Monitoring 提出Deep Researcher Agent，实现零成本监控的深度学习实验全自动框架 large language model ✅
78 Vision-Guided Iterative Refinement for Frontend Code Generation 提出基于视觉反馈迭代优化的前端代码生成框架，提升代码质量。 large language model
79 SemLink: A Semantic-Aware Automated Test Oracle for Hyperlink Verification using Siamese Sentence-BERT 提出SemLink，利用Siamese Sentence-BERT实现高效的语义超链接自动测试。 large language model
80 Label Effects: Shared Heuristic Reliance in Trust Assessment by Humans and LLM-as-a-Judge 揭示人类与LLM信任评估中的标签效应，警惕偏见传播 large language model
81 Foundations for Agentic AI Investigations from the Forensic Analysis of OpenClaw 提出Agentic AI取证分析框架，解决智能体系统取证难题。 large language model
82 On the Role of Fault Localization Context for LLM-Based Program Repair 研究故障定位上下文对基于LLM的程序修复的影响，揭示最佳上下文策略。 large language model
83 LLM Evaluation as Tensor Completion: Low Rank Structure and Semiparametric Efficiency 将LLM评估视为张量补全问题，提出低秩结构和半参数有效性分析方法 large language model
84 MA-IDS: Multi-Agent RAG Framework for IoT Network Intrusion Detection with an Experience Library 提出MA-IDS：一种基于多Agent RAG框架的物联网入侵检测系统，具备经验库。 large language model
85 Your LLM Agent Can Leak Your Data: Data Exfiltration via Backdoored Tool Use 提出Back-Reveal以解决LLM代理数据泄露问题 large language model
86 Towards Effective In-context Cross-domain Knowledge Transfer via Domain-invariant-neurons-based Retrieval 提出DIN-Retrieval，通过领域不变神经元检索实现跨领域知识迁移，提升LLM推理能力。 large language model ✅
87 TFRBench: A Reasoning Benchmark for Evaluating Forecasting Systems TFRBench：用于评估预测系统推理能力的新基准 foundation model
88 TRACE: Capability-Targeted Agentic Training TRACE：面向能力的Agent训练，提升Agent在复杂环境中的任务解决能力 large language model
89 Pressure, What Pressure? Sycophancy Disentanglement in Language Models via Reward Decomposition 通过奖励分解提出新方法以减少语言模型的谄媚行为 large language model

🔬 支柱二：RL算法与架构 (RL & Architecture) (28 篇)

#题目一句话要点标签🔗⭐
90 Scalable and Explainable Learner-Video Interaction Prediction using Multimodal Large Language Models 提出基于多模态大语言模型的可扩展、可解释学习者-视频交互预测方法 predictive model large language model multimodal
91 Hierarchical Semantic Correlation-Aware Masked Autoencoder for Unsupervised Audio-Visual Representation Learning 提出HSC-MAE以解决无监督音视频表示学习中的对齐问题 representation learning masked autoencoder MAE
92 PanLUNA: An Efficient and Robust Query-Unified Multimodal Model for Edge Biosignal Intelligence PanLUNA：一种高效鲁棒的查询统一多模态模型，用于边缘生物信号智能 representation learning foundation model multimodal
93 Structural Rigidity and the 57-Token Predictive Window: A Physical Framework for Inference-Layer Governability in Large Language Models 提出能量基础治理框架以解决大语言模型的可控性问题 world model world models large language model
94 Fusion and Alignment Enhancement with Large Language Models for Tail-item Sequential Recommendation FAERec：融合对齐增强框架，利用LLM提升尾部物品序列推荐效果 contrastive learning curriculum learning large language model
95 Analyzing Symbolic Properties for DRL Agents in Systems and Networking 提出diffRL框架，分析DRL智能体在系统和网络中的符号属性，提升安全部署。 reinforcement learning deep reinforcement learning DRL
96 When Adaptive Rewards Hurt: Causal Probing and the Switching-Stability Dilemma in LLM-Guided LEO Satellite Scheduling 揭示自适应奖励陷阱：因果探测与LLM在LEO卫星调度中的切换-稳定性困境 reinforcement learning deep reinforcement learning DRL
97 Representation learning to advance multi-institutional studies with electronic health record data from US and France 提出基于图的表征学习框架，解决多机构电子病历数据异构性问题。 representation learning large language model
98 RL-Driven Sustainable Land-Use Allocation for the Lake Malawi Basin 提出基于强化学习的土地利用优化框架，用于马拉维湖流域生态系统服务价值最大化。 reinforcement learning deep reinforcement learning PPO
99 Pedagogical Safety in Educational Reinforcement Learning: Formalizing and Detecting Reward Hacking in AI Tutoring Systems 提出教育强化学习中的教学安全框架，并量化AI辅导系统中的奖励利用问题。 reinforcement learning reward design
100 PRAISE: Prefix-Based Rollout Reuse in Agentic Search Training 提出PRAISE框架，通过前缀复用提升Agentic搜索训练效率和奖励分配。 reinforcement learning policy learning large language model
101 A Multi-Agent Reinforcement Learning Framework for Public Health Decision Analysis 提出基于多智能体强化学习的公共卫生决策框架，优化HIV防控资源分配。 reinforcement learning
102 ActionNex: A Virtual Outage Manager for Cloud ActionNex：用于云环境的虚拟故障管理系统，实现端到端故障辅助。 distillation multimodal
103 Decomposing Communication Gain and Delay Cost Under Cross-Timestep Delays in Cooperative Multi-Agent Reinforcement Learning 针对通信延迟的多智能体强化学习，提出CDCMA框架以解耦通信增益与延迟代价 reinforcement learning
104 Comparative reversal learning reveals rigid adaptation in LLMs under non-stationary uncertainty 通过比较逆转学习揭示LLM在非稳态不确定性下的刚性适应 reinforcement learning large language model
105 Search, Do not Guess: Teaching Small Language Models to Be Effective Search Agents 提出轻量级微调方法以提升小型语言模型的搜索能力 distillation large language model
106 Paper Espresso: From Paper Overload to Research Insight Paper Espresso：利用LLM自动发现、总结和分析arXiv趋势论文，助力科研洞察。 reinforcement learning large language model
107 Reflection of Episodes: Learning to Play Game from Expert and Self Experiences 提出基于专家和自我经验反思的ROE框架，解决LLM在复杂星际争霸2环境中的学习问题 reinforcement learning large language model
108 TSPO: Breaking the Double Homogenization Dilemma in Multi-turn Search Policy Optimization TSPO通过引入Turn-level奖励机制，解决多轮搜索策略优化中的双重同质化问题。 reinforcement learning large language model
109 QED-Nano: Teaching a Tiny Model to Prove Hard Theorems QED-Nano：训练小型模型解决奥赛级难题，性能媲美大型闭源模型。 reinforcement learning IMoS
110 Seemingly Simple Planning Problems are Computationally Challenging: The Countdown Game 提出基于倒计时游戏的规划基准，挑战现有LLM辅助规划方法的长程规划能力。 world model world models
111 Neural Assistive Impulses: Synthesizing Exaggerated Motions for Physics-based Characters 提出辅助冲量神经控制，实现物理角色动画中夸张动作的合成 reinforcement learning deep reinforcement learning DRL
112 MARL-GPT: Foundation Model for Multi-Agent Reinforcement Learning MARL-GPT：基于GPT的多智能体强化学习通用模型，实现跨环境任务泛化。 reinforcement learning offline reinforcement learning foundation model
113 Can Large Language Models Reinvent Foundational Algorithms? 利用LLM的Unlearn-and-Reinvent流程，探索其重塑基础算法的能力 reinforcement learning large language model
114 Hierarchical Reinforcement Learning with Augmented Step-Level Transitions for LLM Agents STEP-HRL：增强步级转移的分层强化学习LLM Agent框架 reinforcement learning large language model ✅
115 Can We Trust a Black-box LLM? LLM Untrustworthy Boundary Detection via Bias-Diffusion and Multi-Agent Reinforcement Learning 提出GMRL-BD算法，通过偏差扩散和多智能体强化学习检测LLM不可信边界 reinforcement learning large language model
116 UniCreative: Unifying Long-form Logic and Short-form Sparkle via Reference-Free Reinforcement Learning UniCreative：提出一种无参考强化学习框架，统一长文本逻辑性和短文本创造性。 reinforcement learning
117 Breakthrough the Suboptimal Stable Point in Value-Factorization-Based Multi-Agent Reinforcement Learning 提出多轮价值分解（MRVF）框架，解决多智能体强化学习中价值分解方法易收敛到次优解的问题 reinforcement learning

🔬 支柱一：机器人控制 (Robot Control) (4 篇)

#题目一句话要点标签🔗⭐
118 RAGShield: Detecting Numerical Claim Manipulation in Government RAG Systems RAGShield：检测政府RAG系统中数值声明的篡改，解决嵌入式防御的盲点。 manipulation
119 Automating Cloud Security and Forensics Through a Secure-by-Design Generative AI Framework 提出安全设计生成式AI框架，自动化云安全与取证，提升LLM安全性和取证准确性。 manipulation large language model
120 Receding-Horizon Control via Drifting Models 提出Drifting MPC，结合漂移生成模型与后退 horizon 规划，解决未知动力学下的轨迹优化问题。 MPC trajectory optimization
121 Security in LLM-as-a-Judge: A Comprehensive SoK 首个LLM-as-a-Judge安全综述，揭示潜在风险并探索防御策略。 manipulation

🔬 支柱八：物理动画 (Physics-based Animation) (3 篇)

#题目一句话要点标签🔗⭐
122 Solar-VLM: Multimodal Vision-Language Models for Augmented Solar Power Forecasting 提出Solar-VLM，用于融合多模态信息以增强光伏功率预测。 spatiotemporal large language model multimodal
123 IPSL-AID: Generative Diffusion Models for Climate Downscaling from Global to Regional Scales IPSL-AID：利用生成扩散模型实现全球到区域气候的降尺度，并量化不确定性。 spatiotemporal
124 How AI Coding Agents Modify Code: A Large-Scale Study of GitHub Pull Requests 大规模分析AI代码生成代理的Pull Request，揭示其代码修改模式与人工贡献的差异 diff-sim

🔬 支柱四：生成式动作 (Generative Motion) (2 篇)

#题目一句话要点标签🔗⭐
125 Thinking Diffusion: Penalize and Guide Visual-Grounded Reasoning in Diffusion Multimodal Language Models 提出PSP和VRG，解决扩散多模态语言模型推理中过早生成答案和视觉依赖不足的问题 classifier-free guidance large language model multimodal
126 Hackers or Hallucinators? A Comprehensive Analysis of LLM-Based Automated Penetration Testing 首个LLM驱动的自动化渗透测试框架的系统化知识与大规模评测 penetration large language model

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

#题目一句话要点标签🔗⭐
127 AEGIS: Scaling Long-Sequence Homomorphic Encrypted Transformer Inference via Hybrid Parallelism on Multi-GPU Systems AEGIS：通过多GPU混合并行加速长序列同态加密Transformer推理 OMOMO

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

#题目一句话要点标签🔗⭐
128 Symbolic-Vector Attention Fusion for Collective Intelligence 提出符号-向量注意力融合（SVAF）机制，用于提升集体智能中跨智能体的信息融合效果。 motion representation

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

#题目一句话要点标签🔗⭐
129 VisionClaw: Always-On AI Agents through Smart Glasses VisionClaw：通过智能眼镜实现常时在线的AI Agent egocentric

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

#题目一句话要点标签🔗⭐
130 Learned Elevation Models as a Lightweight Alternative to LiDAR for Radio Environment Map Estimation 提出基于学习的轻量级高程模型，替代LiDAR用于无线电环境地图估计 elevation map

⬅️ 返回 cs.AI 首页 · 🏠 返回主页