cs.AI（2025-03-20）

📊 共 28 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) (21 🔗2) 支柱二：RL算法与架构 (RL & Architecture) (6) 支柱一：机器人控制 (Robot Control) (1)

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

#题目一句话要点标签🔗⭐
1 OmniGeo: Towards a Multimodal Large Language Models for Geospatial Artificial Intelligence 提出OmniGeo，一个用于地理空间人工智能的多模态大语言模型 large language model multimodal instruction following
2 Towards Agentic Recommender Systems in the Era of Multimodal Large Language Models 探索基于多模态大语言模型的Agentic推荐系统，提升推荐的交互性与适应性 large language model multimodal
3 Towards Agentic AI Networking in 6G: A Generative Foundation Model-as-Agent Approach 提出AgentNet框架，利用生成式基础模型赋能6G网络中自主AI Agent的协作。 embodied AI foundation model
4 Large Language Models for Water Distribution Systems Modeling and Decision-Making 提出基于LLM-EPANET架构的框架，用于水分配系统建模和决策，实现自然语言交互。 large language model
5 Echoes of Power: Investigating Geopolitical Bias in US and China Large Language Models 研究揭示中美大型语言模型在回答地缘政治问题时存在的意识形态和文化偏见 large language model
6 Code Evolution Graphs: Understanding Large Language Model Driven Design of Algorithms 提出代码演化图，用于分析LLM驱动的算法设计过程，揭示LLM在进化计算中的代码生成模式。 large language model
7 Bridging Technology and Humanities: Evaluating the Impact of Large Language Models on Social Sciences Research with DeepSeek-R1 利用DeepSeek-R1评估大语言模型对社会科学研究的影响 large language model
8 Using Large Language Models to Categorize Strategic Situations and Decipher Motivations Behind Human Behaviors 利用大型语言模型分类策略情境并解读人类行为动机 large language model
9 Survey on Evaluation of LLM-based Agents 全面评测LLM驱动的智能体：基准、框架与未来方向 generalist agent
10 The Emperor's New Clothes in Benchmarking? A Rigorous Examination of Mitigation Strategies for LLM Benchmark Data Contamination 提出系统评估方法以解决大语言模型基准数据污染问题 large language model ✅
11 Palatable Conceptions of Disembodied Being 探讨具身性缺失的AI系统中意识概念的可能性与哲学挑战 embodied AI
12 Unify and Triumph: Polyglot, Diverse, and Self-Consistent Generation of Unit Tests with LLMs PolyTest：利用多语言和多样性生成自洽的单元测试，显著提升测试质量 large language model
13 Autonomous AI imitators increase diversity in homogeneous information ecosystems 自主AI模仿者在同质化信息生态系统中增加多样性 large language model
14 GAN-enhanced Simulation-driven DNN Testing in Absence of Ground Truth 提出GAN增强的模拟驱动DNN测试方法，解决无真值标签下的测试难题 large language model
15 Advancing Mobile GUI Agents: A Verifier-Driven Approach to Practical Deployment V-Droid：一种基于验证器的移动GUI代理，提升任务自动化性能与效率 large language model ✅
16 DeepPsy-Agent: A Stage-Aware and Deep-Thinking Emotional Support Agent System DeepPsy-Agent：结合心理学三阶段理论与深度学习的情感支持智能体系统 large language model
17 Entropy-based Exploration Conduction for Multi-step Reasoning 提出Entro-duction，通过熵引导LLM进行多步推理的探索深度调整。 large language model
18 Attention Pruning: Automated Fairness Repair of Language Models via Surrogate Simulated Annealing 提出Attention Pruning，通过代理模拟退火自动修复语言模型中的偏见。 large language model
19 ChatGPT and U(X): A Rapid Review on Measuring the User Experience 快速综述ChatGPT用户体验评估方法，填补标准化评估体系的空白。 large language model
20 Detecting LLM-Generated Peer Reviews 提出一种基于隐蔽水印的LLM生成同行评审检测框架，提升检测可靠性。 large language model
21 AutoRedTeamer: Autonomous Red Teaming with Lifelong Attack Integration AutoRedTeamer：基于终身攻击集成的大语言模型自主红队测试框架 large language model

🔬 支柱二：RL算法与架构 (RL & Architecture) (6 篇)

#题目一句话要点标签🔗⭐
22 Deconstructing Long Chain-of-Thought: A Structured Reasoning Optimization Framework for Long CoT Distillation 提出DLCoT框架，优化长链思维蒸馏，提升非同源模型性能。 distillation large language model chain-of-thought
23 Video-VoT-R1: An efficient video inference model integrating image packing and AoE architecture 提出Video-VoT-R1模型，结合图像打包和AoE架构，提升视频语言预训练的推理效率。 reinforcement learning large language model multimodal
24 Diffusion-augmented Graph Contrastive Learning for Collaborative Filter 提出DGCL：一种扩散增强的图对比学习协同过滤方法，缓解数据稀疏性问题。 representation learning contrastive learning
25 Towards Automated Semantic Interpretability in Reinforcement Learning via Vision-Language Models 提出iTRACE，利用视觉-语言模型自动构建可解释强化学习策略。 reinforcement learning
26 Reinforcement Learning-based Heuristics to Guide Domain-Independent Dynamic Programming 提出基于强化学习的启发式方法，指导领域无关动态规划搜索 reinforcement learning
27 Unreal-MAP: Unreal-Engine-Based General Platform for Multi-Agent Reinforcement Learning 提出基于虚幻引擎的通用多智能体强化学习平台Unreal-MAP，促进算法与定制任务的集成。 reinforcement learning

🔬 支柱一：机器人控制 (Robot Control) (1 篇)

#题目一句话要点标签🔗⭐
28 Real AI Agents with Fake Memories: Fatal Context Manipulation Attacks on Web3 Agents 提出上下文操控攻击以揭示Web3智能代理的安全漏洞 manipulation

⬅️ 返回 cs.AI 首页 · 🏠 返回主页