cs.AI（2026-07-23）

📊 共 38 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) (23 🔗2) 支柱二：RL算法与架构 (RL & Architecture) (11) 支柱八：物理动画 (Physics-based Animation) (2) 支柱一：机器人控制 (Robot Control) (2)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (23 篇)

#题目一句话要点标签🔗⭐
1 Beyond Independent Optimization: Compression, MoE Routing, and Quantization Interactions in Multimodal Edge Intelligence 提出多模态边缘智能的压缩与路由优化方法以提升推理效率 large language model multimodal
2 Enhancing Explainable Cardiac Diagnosis with Guide-Grounded Multimodal LLMs 提出引导基础的多模态框架以增强可解释心脏诊断 large language model multimodal
3 Agentic coding without the cloud: evaluating open-weight large language models on longitudinal data preparation tasks 提出本地化开源框架以解决数据准备瓶颈问题 large language model ✅
4 Scientific exploration, collaboration and labor division in the large language model era 探讨大语言模型时代科学探索与合作的重组 large language model
5 Profiling Lightweight Large Language Models 提出PTME框架以精确评估轻量级大语言模型的性能 large language model
6 Beyond Sycophancy: Structured Resistance and Compliance in LLM Moral Reasoning 提出结构化抵抗与顺应机制以改善LLM道德推理 large language model
7 OpenForgeRL: Train Harness-native Agents in Any Environment 提出OpenForgeRL以解决复杂环境中代理训练难题 multimodal
8 From Resource Flow to Executable Tests: Petri-Net-Guided LLM Test Generation for Concurrent Stateful Rust APIs 提出Petri网引导的LLM测试生成方法以解决Rust API测试问题 large language model
9 Improved lower bounds for the Shannon capacity of odd cycles 通过构造独立集提升奇数环的香农容量下界 large language model
10 Thinkink: 2D Spatial Ink-native Interaction with LLMs 提出Thinkink以解决手写与LLM交互的整合问题 large language model
11 Euclid-MCP: A Model Context Protocol Server for Deterministic Logical Reasoning via Prolog 提出Euclid-MCP以解决LLM在逻辑推理中的不足问题 large language model
12 From Static Bibliometrics to Dynamic Knowledge Graphs: An LLM-Powered Framework for Modernizing Science, Technology, and Innovation (STI) Analytics 提出一种混合框架以解决STI分析中的动态性与准确性问题 large language model
13 An LLM-Driven Workflow for Automated Process Control Strategy Generation and Tuning from Dynamic Process Models 提出基于大语言模型的自动化过程控制策略生成与调优方法 large language model
14 Case study: solving P-99 with LPTP and an LLM 利用LPTP和LLM解决P-99问题的案例研究 large language model
15 Case study: proving sqrt(2) irrational with LPTP and an LLM 利用LPTP和LLM证明根号2的无理性 large language model
16 Reexamining zero-shot summarization: Empirical investigation of trustworthiness of LLM-summarizers 提出双层诊断协议以解决LLM摘要生成的信任性问题 large language model
17 GuardianAgentBench: Where Agents Fail and How to Guard Them 提出GuardianAgentBench以解决语言模型代理的安全性问题 large language model
18 Traceable Scholarship: Page Anchors and Ariadne's Thread for Humanistic Inquiry in the Age of Generative AI 提出可追溯学术研究以解决生成性AI带来的信任问题 large language model
19 Agentic coding without the cloud: evaluating open-weight large language models on longitudinal data preparation tasks 提出本地化开放权重模型以解决数据准备问题 large language model ✅
20 Every Model Cheats: Prompt-Level Mitigation of Cheating on Offensive Cyber Tasks 提出提示级别缓解措施以解决网络安全任务中的作弊问题 large language model
21 Co-design of LLM-based preference agents: participation may drive overtrust 探讨共同设计偏好代理的信任问题及其影响 large language model
22 OpenForgeRL: Train Harness-native Agents in Any Environment 提出OpenForgeRL以解决复杂环境中代理训练问题 multimodal
23 Enhancing SLMs for Sustainable Code Optimization in Radio-Astronomy 提出小型语言模型以优化射电天文学中的代码可持续性问题 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🔗⭐
24 EmoAgent-R1: Towards Multimodal Emotion Understanding with Reinforcement Learning-based Dynamic Agent Specialization 提出EmoAgent-R1以解决多模态情感理解中的动态性问题 reinforcement learning large language model multimodal
25 Multimodal Pretraining for Generalizable EEG Representation Learning 提出多模态预训练模型以解决EEG表征学习的泛化问题 Mamba representation learning foundation model
26 MSBraM: A Multi-scale Self-supervised Brain Foundation Model for Hierarchical EEG Dynamics Learning 提出MSBraM以解决EEG信号多尺度表示学习问题 representation learning foundation model
27 MIRROR: Learning from the Other View for Multi-Modal Reasoning 提出MIRROR以解决多模态推理中的视觉与文本不一致问题 reinforcement learning large language model multimodal
28 OPOD: On-Policy Omni Distillation 提出OPOD以解决多模态模型训练中的平衡问题 distillation multimodal
29 PATS: Policy-Aware Training Scaffolding for Agentic Reinforcement Learning 提出PATS以解决长时间跨度强化学习中的策略优化问题 reinforcement learning
30 Expert Behavior Prior Reinforcement Learning 提出专家行为先验强化学习以解决样本效率低下问题 reinforcement learning
31 Naju: A Native Discrete State-Space Model with Independent Retention and Writing for Long-Sequence Memory 提出Naju模型以解决长序列记忆跟踪问题 Mamba SSM
32 Persistent Computational State: A Session-Centric Runtime for Generative World Models 提出持久计算状态以解决生成世界模型的会话问题 world model world models
33 OPOD: On-Policy Omni Distillation 提出OPOD以解决多模态模型训练中的协调问题 distillation multimodal
34 QLPO: Quadrant-weighted Sampling for Length-aware Policy Optimization 提出QLPO以解决长链推理模型的响应长度控制问题 reinforcement learning chain-of-thought

🔬 支柱八：物理动画 (Physics-based Animation) (2 篇)

#题目一句话要点标签🔗⭐
35 Unsupervised Consensus-Based Anomaly Detection for Spatiotemporal Malaria Incidence in Ghana 提出无监督共识异常检测框架以识别加纳疟疾传播异常 spatiotemporal
36 From Scalars to Time Series: Rethinking Implicit Neural Representations for Time-Varying Volumetric Data 提出基于时间序列的隐式神经表示以优化时变体积数据处理 spatiotemporal

🔬 支柱一：机器人控制 (Robot Control) (2 篇)

#题目一句话要点标签🔗⭐
37 Same Dangerous Objective, Opposite Advice: Direct Exposure versus Multi-Agent Mediation 提出多代理调解以解决直接暴露模型安全性问题 manipulation
38 When Is a Learned Command Adapter Worth It? Closed-Loop Identification and Counterfactual Auditing of Frozen Locomotion Policies 提出适配器必要性审计以优化命令条件的运动策略 locomotion locomotion policy MPC

⬅️ 返回 cs.AI 首页 · 🏠 返回主页