cs.AI（2025-08-19）

📊 共 24 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) (15 🔗3) 支柱二：RL算法与架构 (RL & Architecture) (5 🔗2) 支柱一：机器人控制 (Robot Control) (2) 支柱三：空间感知与语义 (Perception & Semantics) (1) 支柱八：物理动画 (Physics-based Animation) (1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (15 篇)

#题目一句话要点标签🔗⭐
1 CrafterDojo: A Suite of Foundation Models for Building Open-Ended Embodied Agents in Crafter 提出CrafterDojo以解决通用体智能体研究的快速原型问题 foundation model instruction following
2 Neuro-Symbolic Artificial Intelligence: Towards Improving the Reasoning Abilities of Large Language Models 提出神经符号人工智能以提升大型语言模型的推理能力 large language model ✅
3 Large Language Models are Highly Aligned with Human Ratings of Emotional Stimuli 探讨大型语言模型与人类情感评估的高度一致性 large language model
4 UniECS: Unified Multimodal E-Commerce Search Framework with Gated Cross-modal Fusion 提出UniECS以解决电商多模态检索系统的局限性 multimodal ✅
5 COMPASS: A Multi-Dimensional Benchmark for Evaluating Code Generation in Large Language Models 提出COMPASS以解决代码生成评估的多维度问题 large language model
6 Equinox: Holistic Fair Scheduling in Serving Large Language Models 提出Equinox以解决大语言模型服务中的公平调度问题 large language model
7 InPars+: Supercharging Synthetic Data Generation for Information Retrieval Systems 提出InPars+以提升神经信息检索系统的合成数据生成 large language model chain-of-thought ✅
8 Explaining Hitori Puzzles: Neurosymbolic Proof Staging for Sequential Decisions 提出神经符号方法以解释Hitori谜题的决策过程 large language model
9 Incident Analysis for AI Agents 提出AI代理事件分析框架以解决安全隐患问题 chain-of-thought
10 ChronoLLM: Customizing Language Models for Physics-Based Simulation Code Generation 提出ChronoLLM以定制语言模型生成物理仿真代码 large language model
11 Prompt Orchestration Markup Language 提出POML以解决大型语言模型提示结构与集成问题 large language model
12 The Collaboration Paradox: Why Generative AI Requires Both Strategic Intelligence and Operational Stability in Supply Chain Management 提出协作悖论以解决供应链管理中的AI行为问题 large language model
13 Agentic DraCor and the Art of Docstring Engineering: Evaluating MCP-empowered LLM Usage of the DraCor API 提出MCP服务器以优化LLM与DraCor API的交互 large language model
14 The Hidden Cost of Readability: How Code Formatting Silently Consumes Your LLM Budget 提出代码格式优化策略以降低LLM计算成本 large language model
15 CCFC: Core & Core-Full-Core Dual-Track Defense for LLM Jailbreak Protection 提出CCFC框架以解决大型语言模型的越狱攻击问题 large language model

🔬 支柱二：RL算法与架构 (RL & Architecture) (5 篇)

#题目一句话要点标签🔗⭐
16 Breaking the SFT Plateau: Multimodal Structured Reinforcement Learning for Chart-to-Code Generation 提出多模态结构化强化学习以解决图表到代码生成问题 reinforcement learning multimodal
17 MHSNet:An MoE-based Hierarchical Semantic Representation Network for Accurate Duplicate Resume Detection with Large Language Model 提出MHSNet以解决简历重复检测问题 contrastive learning large language model
18 Toward Better EHR Reasoning in LLMs: Reinforcement Learning with Expert Attention Guidance 提出EAG-RL以解决LLMs在EHR推理中的不足问题 reinforcement learning large language model ✅
19 ComputerRL: Scaling End-to-End Online Reinforcement Learning for Computer Use Agents 提出ComputerRL以解决桌面智能代理的训练效率问题 reinforcement learning PULSE ✅
20 CausalPlan: Empowering Efficient LLM Multi-Agent Collaboration Through Causality-Driven Planning 提出CausalPlan以解决LLM多智能体协作中的因果推理问题 reinforcement learning large language model

🔬 支柱一：机器人控制 (Robot Control) (2 篇)

#题目一句话要点标签🔗⭐
21 BiND: A Neural Discriminator-Decoder for Accurate Bimanual Trajectory Prediction in Brain-Computer Interfaces 提出BiND以解决脑机接口中双手轨迹预测问题 bi-manual
22 CIA+TA Risk Assessment for AI Reasoning Vulnerabilities 提出CIA+TA框架以解决AI推理脆弱性问题 manipulation penetration

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

#题目一句话要点标签🔗⭐
23 DegDiT: Controllable Audio Generation with Dynamic Event Graph Guided Diffusion Transformer 提出DegDiT以解决可控音频生成中的时间定位与效率问题 open-vocabulary open vocabulary TAMP

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

#题目一句话要点标签🔗⭐
24 Discrete Optimization of Min-Max Violation and its Applications Across Computational Sciences 提出离散最小-最大违例优化以解决多领域问题 PULSE

⬅️ 返回 cs.AI 首页 · 🏠 返回主页