cs.AI（2026-08-06）

📊 共 45 篇论文 | 🔗 7 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (21 🔗4) 支柱二：RL算法与架构 (RL & Architecture) (20 🔗3) 支柱三：空间感知与语义 (Perception & Semantics) (2) 支柱一：机器人控制 (Robot Control) (2)

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

#题目一句话要点标签🔗⭐
1 Seeing Is Not Deciding: Can Multimodal LLMs Act as Effective CEOs? 提出C-SUITEBENCH以解决多模态决策中的信息整合问题 large language model multimodal visual grounding
2 F$^2$Agent: Financial Fusion of Agentic Intelligence for Multimodal Trading 提出F$^2$Agent以解决多模态金融交易中的信息融合问题 large language model multimodal
3 Reducing belief in conspiracy theories as they unfold using large language models 利用大型语言模型减少对阴谋论的信任 large language model
4 Poli-Bias: Understanding and Measuring Large Language Model Biases in International Political Conflicts 提出Poli-Bias框架以测量大型语言模型中的政治偏见 large language model
5 Evaluating Investment Logic in Large Language Models: A Real-World Benchmark Towards Personalzied Financial Agents 提出InvestLogicBench以评估个性化金融代理的投资逻辑 large language model
6 The em-dash em-beds in Congress: A population-level rise in em-dash frequency in U.S. congressional press releases at the dawn of the large-language-model era, 2021-2025 分析美国国会新闻稿中无间隔破折号频率的变化 large language model
7 Unified Agent: Managing Interactions across Devices 提出统一代理以解决跨设备交互管理问题 large language model multimodal
8 Bias Analysis of L2 Speaking Assessment Systems Using Concept Activation Vectors 基于概念激活向量的L2口语评估系统偏差分析 foundation model multimodal
9 TS-RAG: Retrieval Augmented Generation for Time Series Forecasting 提出TS-RAG以解决时间序列预测中的数据不足问题 large language model
10 The Illusion of Visual Tool-Use: A Causal Audit of Thinking with Images 提出因果审计方法以解决视觉工具使用的有效性问题 multimodal ✅
11 What Current AI Benchmarks Leave Unmeasured: Modality, Search, Citations, and Implications (for Safety Evaluations) 审计AI基准评估中的多模态与一致性问题 large language model
12 Audio-to-Score Transcription using Pre-trained Features, Data Augmentation, and the New SheetSage-A2S Dataset 提出SheetSage-A2S数据集以解决流行音乐音频转谱问题 multimodal ✅
13 Learning Globally Reusable Skills for Coding Agents 提出全球可重用技能演化框架GSE以解决编码代理技能泛化问题 large language model
14 Mind the Gaps: Mixture-of-Minds for Human Simulation 提出Anacreon以解决个体层面的人类模拟问题 large language model
15 BALANCE: Hybrid Autoregressive-Speculative LLM Inference in Wireless Edge Networks 提出BALANCE框架以解决边缘网络中LLM推理的延迟与内存问题 large language model
16 A Two-Tier Perspective on Inference-Time Parallelism in Multi-Agent LLM Systems 提出TIPEX框架以提升多智能体LLM系统推理效率 large language model
17 Measuring and Detecting Harmful AI Sycophancy 提出CAP框架以自动检测有害的AI谄媚行为 large language model
18 Hyper-ES: Effective Evolution Strategies for LLM Reasoning via Descent Direction Merging 提出Hyper-ES以解决大规模语言模型推理中的优化效率问题 large language model ✅
19 Characterizing the Quality Profile of AI-Generated C++ in Production 研究AI生成C++代码质量以提升生产效率 large language model
20 CyberForge: Verified Vulnerability Injection at Repository Level for Cybersecurity Agent Training 提出CyberForge以解决网络安全代理训练数据不足问题 large language model
21 Audio-to-Score Transcription using Pre-trained Features, Data Augmentation, and the New SheetSage-A2S Dataset 提出SheetSage-A2S数据集以解决流行音乐音频转谱问题 multimodal ✅

🔬 支柱二：RL算法与架构 (RL & Architecture) (20 篇)

#题目一句话要点标签🔗⭐
22 EnvACE: Internalizing Environment Dynamics via World Rehearsal for Agentic Reinforcement Learning 提出EnvACE以解决长时间工具使用的环境交互问题 reinforcement learning policy learning world model ✅
23 DreamGuard: Efficient Runtime Guardrail for LLM Agents via Risk-Aware World Model 提出DreamGuard以解决LLM代理的风险管理问题 world model world models large language model
24 AgentOPSD: Recursive Self-Distillation for Agentic Reinforcement Learning 提出AgentOPSD以解决长时间多回合任务中的信用分配问题 reinforcement learning teacher-student distillation
25 AppDeltaWorld: Transition-Grounded Delta Code World Model for Mobile GUI Agents 提出AppDeltaWorld以解决移动GUI代理的环境建模问题 reinforcement learning world model world models
26 ViSR-KGC: Visual Subgraph Reasoning with Vision-Language Models for Multimodal Knowledge Graph Completion 提出ViSR-KGC以解决多模态知识图谱补全问题 representation learning multimodal
27 When Agentic AI Meets Integrated Sensing and Communication 提出AISAC框架以整合智能感知与通信技术 reinforcement learning world model world models
28 GAUGE: A Measurement-Grounded Benchmark for Physical Fidelity in Simulation Engines and Video World Models 提出GAUGE基准以评估模拟引擎和视频世界模型的物理真实度 world model world models
29 DASH: Divergence-Adaptive Supervision Horizons for On-Policy Self-Distillation of Reasoning Models 提出DASH以解决标准OPSD在时间结构利用上的不足 reinforcement learning distillation large language model ✅
30 iARCS: Iterative Agentic RL for Controllable 3D Scene Generation 提出iARCS框架以解决3D场景生成中的功能约束问题 reinforcement learning traversability embodied AI
31 StepReflect: Structured UI Transition Reflection for Mobile GUI Agents 提出StepReflect以解决移动GUI代理的准确动作反映问题 teacher-student distillation multimodal
32 From Passive Mirrors to Active Agents: Holonic Digital Twins for Physical AI over Networks 提出HDT-Nets框架以解决物理AI协调问题 world model world models spatiotemporal
33 Training a Conditioned Video Game Agent on a VLM Annotated Dataset 提出基于视觉语言模型的强化学习视频游戏代理训练方法 reinforcement learning policy learning offline RL
34 Contextual Information Policy Optimization for Search Agents 提出上下文信息策略优化以解决搜索代理的推理问题 reinforcement learning large language model
35 Subliminal Learning is Non-Semantic Distillation 提出隐性学习以解决AI系统可预测性问题 distillation
36 VLMs for Videogame Data Annotation 利用视觉语言模型进行视频游戏数据标注以提升训练效果 reinforcement learning offline reinforcement learning
37 TaskSense: Focusing on What Matters in World Models 提出TaskSense以解决视觉控制中的任务相关性问题 world model world models dreamer
38 DASH: Divergence-Adaptive Supervision Horizons for On-Policy Self-Distillation of Reasoning Models 提出DASH以解决标准OPSD在时间结构利用上的不足 reinforcement learning distillation large language model ✅
39 Shape Your Feed: An LLM-based Agentic System for Conversational Recommendation 提出基于LLM的SYF系统以解决推荐系统用户偏好表达不足问题 DPO direct preference optimization multimodal
40 WebGrader: Training LLMs for Web Development with Self-Evolving Programmatic Grader 提出WebGrader以解决大语言模型在网页开发中的功能缺口问题 reinforcement learning reward design large language model
41 Contextual Information Policy Optimization for Search Agents 提出上下文信息策略优化以解决搜索代理的推理问题 reinforcement learning large language model

🔬 支柱三：空间感知与语义 (Perception & Semantics) (2 篇)

#题目一句话要点标签🔗⭐
42 GSBF: Gaussian Splatting for Environment-Aware Beamforming 提出GSBF以解决MIMO通信中的波束形成问题 3D gaussian splatting gaussian splatting splatting
43 CogVis: Must Open-Vocabulary Change Detection Perceive the Scene Anew for Every Query? 提出CogVis以解决开放词汇变化检测中的场景感知问题 open-vocabulary open vocabulary

🔬 支柱一：机器人控制 (Robot Control) (2 篇)

#题目一句话要点标签🔗⭐
44 From Economic Agents to Agentic Economies: A Systems Blueprint for Economic World Models 提出经济世界模型的实施蓝图以推动经济模拟环境的发展 sim-to-real world model world models
45 Resourced Authority A Mechanism-Design Model for Participatory Governance of Deployed AI Agents 提出资源授权机制设计模型以实现AI代理的参与式治理 manipulation

⬅️ 返回 cs.AI 首页 · 🏠 返回主页