cs.AI（2026-10-20）

📊 共 41 篇论文 | 🔗 6 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (23 🔗3) 支柱二：RL算法与架构 (RL & Architecture) (13 🔗2) 支柱一：机器人控制 (Robot Control) (3 🔗1) 支柱七：动作重定向 (Motion Retargeting) (1) 支柱八：物理动画 (Physics-based Animation) (1)

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

#题目一句话要点标签🔗⭐
1 IECG: Weakly-Supervised Interpretability-Enhancing Foundation Model for Electrocardiogram 提出IECG以解决心电图分析中的可解释性问题 foundation model multimodal
2 TriGeoBench: A Multimodal Mathematical Geometry Benchmark with Figures and TikZ 提出TriGeoBench以解决几何推理不足的问题 large language model multimodal
3 Beyond the Sequentiality Bottleneck: A Hybrid Diffusion-Autoregressive Framework for Efficient Multimodal Reasoning 提出CoRAL框架以解决自回归模型推理效率低下问题 multimodal
4 ACE-Compression: An Ability-Centric Evaluation of Compressed Large Language Models 提出ACE-Compression框架以评估压缩大语言模型的能力保留 large language model
5 MMEnterprise-Bench: Evaluating Enterprise Agents over Evolving Multimodal Records 提出MMEnterprise-Bench以评估企业智能体在多模态记录中的表现 multimodal ✅
6 Hallucinations of Large Language Models in Cross-Language Code Translation: Phenomena and Mitigation 提出HaluCodeTran以缓解跨语言代码翻译中的幻觉问题 large language model
7 A Unified Attack Framework against Statistical Generative Watermarks for Large Language Models 提出统一攻击框架以破解大型语言模型的统计生成水印 large language model
8 ChartEvo: Self-Evolving Agent Harnesses for Long-Tail Chart Reasoning 提出ChartEvo以解决长尾图表推理问题 large language model multimodal
9 DIVE: Benchmarking Full-Duplex Interaction with Visual Engagement in Continuous Streams 提出DIVE基准以解决实时多模态交互评估问题 large language model multimodal
10 KPHarness: Co-Evolving Knowledge and Probing for Long-Horizon Gaming Harnesses 提出KPHarness框架以解决长时间游戏中的干预适应问题 vision-language-action
11 Local Work, Global Progress: A Convergence Theory of AI Agent Collaboration 提出基于块优化的AI代理协作理论以提升任务效率 large language model
12 From Code Generator to Engineering Companion: A Plant-aware Agent for PLC Programming 提出植物感知对话助手以优化PLC编程过程 large language model
13 Beyond Scalar Leaderboards: Adaptive Sampling for Pareto Frontier Identification in Multi-Objective LLM Arenas 提出VB-EGE算法以高效识别多目标LLM的Pareto前沿 large language model
14 GBClaw: An agentic AI framework for atomistic grain boundary research and mechanistic discovery 提出GBClaw框架以解决晶界研究中的复杂性问题 large language model
15 Semantic Observability for Autonomous Systems Diagnostics 提出语义可观测性框架以解决故障诊断问题 large language model
16 Stratified Sparse Attention: Reducing the Sparse-to-Dense Gap at Extreme Sparsity 提出分层稀疏注意力以解决极端稀疏下的性能瓶颈问题 large language model ✅
17 A Digital Twin to Improve Cyber-Social Impact 提出数字双胞胎框架以改善网络社会影响评估 large language model
18 Local Inference of LLMs on Long Contexts: A Data Systems Perspective 提出本地推理方法以解决长上下文模型的使用限制 large language model
19 MaD-RSI: Model-as-Data for Recursive Self-Improvement of Agent Harnesses 提出MaD-RSI以解决大语言模型自我改进的局限性 large language model
20 RAG-NIMBF: A Non-Invasive, Modular Benchmarking Framework for Automated RAG Configuration Analysis 提出RAG-NIMBF以解决RAG配置分析的挑战 large language model
21 E8-QAT: Efficient 2-Bit Quantization-Aware Training with Implicit Lattice Codebooks 提出E8-QAT以解决低比特量化训练中的精度保持问题 large language model
22 Pattern Intelligence: Unsupervised, LLM-Interpreted Discovery of Systematic Underpayment and Denial Patterns in Healthcare Revenue Cycle Management 提出Pattern Intelligence以解决医疗收入周期管理中的系统性欠款与拒赔问题 large language model
23 When Can LLM Reasoning Reliably Reuse Intermediate Results? 提出中间结果重用策略以提高大语言模型推理效率 large language model ✅

🔬 支柱二：RL算法与架构 (RL & Architecture) (13 篇)

#题目一句话要点标签🔗⭐
24 Staged Continual Adaptation of Multimodal Foundation Models for Japanese Financial Documents 提出分阶段持续适应方法以优化日本金融文档的多模态基础模型 distillation foundation model multimodal
25 Slot-JEPA: Object-Structured Prediction Interfaces for Latent World Models 提出Slot-JEPA以解决现有JEPAs在对象结构预测中的不足 world model world models JEPA
26 Quantum-JEPA: Quantum Born Priors for Latent-Action JEPA World Models 提出Quantum-JEPA以解决世界建模中的不确定性问题 world model world models JEPA
27 HypTI: Hyperbolic Geometry-Aware Multimodal Feature Fusion Framework for Dementia Detection Based on Tabular-Image Data 提出HypTI框架以解决多模态数据融合在痴呆检测中的问题 representation learning multimodal
28 Reusing Multimodal Agents: Efficient Evaluation of Sound Designs in Games 提出多模态学习框架以高效评估游戏音效设计 reinforcement learning multimodal ✅
29 Spatially Informed and Direction Sensitive Representation Learning for Spatial Transcriptomics 提出Tessera框架以解决空间转录组学中的表示学习问题 representation learning foundation model
30 TrajEvolve: Trajectory Evolution for Reinforcement Learning with Hindsight Credit Assignment 提出TrajEvolve以解决长时间任务中的轨迹优化问题 reinforcement learning large language model
31 Robust Representation Learning for Spiking Neural Networks based on Conditional Information Bottleneck 提出SCIB以解决脉冲神经网络的鲁棒性问题 representation learning spatiotemporal
32 SAPD: Learning from Agent Experience without Environment Interaction 提出SAPD以解决无环境交互的经验学习问题 reinforcement learning distillation large language model
33 Interruption-Aware Recurrent Reinforcement Learning for UAS Navigation Under Temporary Overrides 提出中断感知递归强化学习以解决UAS导航中的临时覆盖问题 reinforcement learning
34 Building Telecom Language Models with RSOD: Domain Data, Specialist Post-Training, Memory-Conscious Distillation, and Networking Evaluation 提出RSOD框架以构建电信语言模型，提升网络知识与专业行为的结合 distillation
35 FALCA: Falsification-Aware Reinforcement Learning with Hierarchical Credit Assignment for Long-Horizon Deep Search Agents 提出FALCA框架以解决长时间深度搜索中的错误累积问题 reinforcement learning
36 Ask to Distill: Learning to Synthesize Data for On-Policy Distillation 提出Ask-to-Distill框架以优化问题生成用于蒸馏训练 distillation ✅

🔬 支柱一：机器人控制 (Robot Control) (3 篇)

#题目一句话要点标签🔗⭐
37 Exploring Temporal Context-Aware Physical Grounding in Latent World Models 提出$ ext{G}$-JEPA以解决JEPA模型中的时间上下文聚合问题 manipulation world model world models ✅
38 When LLMs Manipulate: Measuring Behavioral, Social, and Temporal Effects 提出行为计算框架以量化LLM的操控影响 manipulation large language model
39 MAS-Guard: Unified Guardrails from Individual Agents to Multi-Agent LLM Systems 提出MAS-Guard以解决多代理LLM系统的安全问题 manipulation large language model

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

#题目一句话要点标签🔗⭐
40 From Encoding to Generation: Towards Empathetic Large Language Models 提出情感理解与生成框架以提升语言模型的同理心表现 motion representation large language model

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

#题目一句话要点标签🔗⭐
41 When Does Feedback Help? Auditing Learned Voltage-Clamp Experimental Design 审计学习电压钳实验设计中的反馈作用 PULSE

⬅️ 返回 cs.AI 首页 · 🏠 返回主页