cs.AI（2025-08-26）

📊 共 41 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) (29 🔗4) 支柱二：RL算法与架构 (RL & Architecture) (10 🔗3) 支柱五：交互与反应 (Interaction & Reaction) (1) 支柱八：物理动画 (Physics-based Animation) (1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (29 篇)

#题目一句话要点标签🔗⭐
1 CAC-CoT: Connector-Aware Compact Chain-of-Thought for Efficient Reasoning Data Synthesis Across Dual-System Cognitive Tasks 提出CAC-CoT以提升双系统认知任务中的推理效率 large language model chain-of-thought
2 MedVQA-TREE: A Multimodal Reasoning and Retrieval Framework for Sarcopenia Prediction 提出MedVQA-TREE框架以解决肌肉减少症预测问题 multimodal
3 Enabling Transparent Cyber Threat Intelligence Combining Large Language Models and Domain Ontologies 提出结合本体论与大语言模型的网络威胁情报方法以解决信息提取问题 large language model
4 Federated Fine-Tuning of Sparsely-Activated Large Language Models on Resource-Constrained Devices 提出FLUX以解决资源受限设备上MoE模型的联邦微调问题 large language model
5 Investigating Advanced Reasoning of Large Language Models via Black-Box Interaction 提出黑箱交互评估范式以提升大语言模型推理能力 large language model
6 Interactive Evaluation of Large Language Models for Multi-Requirement Software Engineering Tasks 提出交互式评估框架以提升大语言模型在软件工程任务中的表现 large language model
7 eSkinHealth: A Multimodal Dataset for Neglected Tropical Skin Diseases 提出eSkinHealth数据集以解决皮肤忽视热带疾病数据稀缺问题 multimodal
8 "She was useful, but a bit too optimistic": Augmenting Design with Interactive Virtual Personas 提出交互式虚拟角色以解决传统用户画像的局限性 large language model multimodal
9 QAgent: An LLM-based Multi-Agent System for Autonomous OpenQASM programming 提出QAgent以解决OpenQASM编程自动化问题 large language model chain-of-thought
10 VistaWise: Building Cost-Effective Agent with Cross-Modal Knowledge Graph for Minecraft 提出VistaWise以解决Minecraft中知识缺乏问题 large language model multimodal
11 ArgRAG: Explainable Retrieval Augmented Generation using Quantitative Bipolar Argumentation 提出ArgRAG以解决RAG在高风险领域的决策透明性问题 large language model
12 VISION: Robust and Interpretable Code Vulnerability Detection Leveraging Counterfactual Augmentation 提出VISION框架以解决代码漏洞检测中的虚假相关性问题 large language model
13 FALCON: Autonomous Cyber Threat Intelligence Mining with LLMs for IDS Rule Generation 提出FALCON以实现自主生成入侵检测系统规则 large language model
14 Quantized but Deceptive? A Multi-Dimensional Truthfulness Evaluation of Quantized LLMs 提出TruthfulnessEval框架以评估量化LLM的真实性问题 large language model
15 Beyond Memorization: Reasoning-Driven Synthesis as a Mitigation Strategy Against Benchmark Contamination 提出基于推理驱动合成的策略以应对基准污染问题 large language model
16 An Investigation on Group Query Hallucination Attacks 提出群查询攻击以揭示大语言模型的潜在缺陷 large language model
17 Can Structured Templates Facilitate LLMs in Tackling Harder Tasks? : An Exploration of Scaling Laws by Difficulty 提出结构化解决方案模板以提升LLMs在复杂任务中的表现 large language model
18 A Concurrent Modular Agent: Framework for Autonomous LLM Agents 提出并实现了并发模块代理框架以解决自主LLM代理的协调问题 large language model ✅
19 AI Models Exceed Individual Human Accuracy in Predicting Everyday Social Norms 提出大型语言模型以超越人类预测社会规范的能力 large language model
20 Enabling MoE on the Edge via Importance-Driven Expert Scheduling 通过重要性驱动的专家调度实现边缘设备上的MoE large language model
21 Novel Approaches to Artificial Intelligence Development Based on the Nearest Neighbor Method 基于最近邻方法提出新型人工智能开发方案以解决神经网络局限性 large language model
22 Judicial Requirements for Generative AI in Legal Reasoning 提出司法AI系统核心能力以提升法律推理可靠性 large language model
23 ClusterFusion: Expanding Operator Fusion Scope for LLM Inference via Cluster-Level Collective Primitive 提出ClusterFusion以解决LLM推理中的延迟问题 large language model ✅
24 CausalMACE: Causality Empowered Multi-Agents in Minecraft Cooperative Tasks 提出CausalMACE以解决Minecraft多智能体协作任务中的因果依赖问题 large language model
25 Insights into User Interface Innovations from a Design Thinking Workshop at deRSE25 通过设计思维工作坊提出LLM用户界面创新 large language model
26 Bias Mitigation Agent: Optimizing Source Selection for Fair and Balanced Knowledge Retrieval 提出偏差缓解代理以优化知识检索中的源选择 large language model
27 AppAgent-Pro: A Proactive GUI Agent System for Multidomain Information Integration and User Assistance 提出AppAgent-Pro以解决信息获取的被动性问题 large language model ✅
28 Beyond Benchmark: LLMs Evaluation with an Anthropomorphic and Value-oriented Roadmap 提出人性化与价值导向的评估框架以解决LLMs评估不足问题 large language model ✅
29 A Case Study on the Effectiveness of LLMs in Verification with Proof Assistants 研究大型语言模型在证明助手中的验证效果 large language model

🔬 支柱二：RL算法与架构 (RL & Architecture) (10 篇)

#题目一句话要点标签🔗⭐
30 PRISM: Robust VLM Alignment with Principled Reasoning for Integrated Safety in Multimodality 提出PRISM以解决视觉语言模型安全性问题 DPO direct preference optimization multimodal ✅
31 EMind: A Foundation Model for Multi-task Electromagnetic Signals Understanding 提出EMind以解决电磁信号理解的多任务挑战 representation learning foundation model ✅
32 Reasoning LLMs in the Medical Domain: A Literature Survey 调查大型语言模型在医疗领域推理能力的演变与应用 reinforcement learning large language model multimodal
33 RLMR: Reinforcement Learning with Mixed Rewards for Creative Writing 提出RLMR以解决创意写作中的主观与客观平衡问题 reinforcement learning large language model instruction following
34 STARec: An Efficient Agent Framework for Recommender Systems via Autonomous Deliberate Reasoning 提出STARec以解决推荐系统中的静态用户建模问题 distillation reward shaping large language model
35 FormaRL: Enhancing Autoformalization with no Labeled Data 提出FormaRL以解决自动形式化数据稀缺问题 reinforcement learning large language model ✅
36 A Survey on Cloud-Edge-Terminal Collaborative Intelligence in AIoT Networks 提出云-边-终端协同智能以解决AIoT网络中的分布式计算问题 reinforcement learning large language model
37 MUA-RL: Multi-turn User-interacting Agent Reinforcement Learning for agentic tool use 提出MUA-RL以解决多轮用户交互中的工具使用问题 reinforcement learning
38 Playstyle and Artificial Intelligence: An Initial Blueprint Through the Lens of Video Games 提出游戏风格作为智能体决策行为的新视角 reinforcement learning imitation learning
39 Attackers Strike Back? Not Anymore -- An Ensemble of RL Defenders Awakens for APT Detection 提出一种新框架以解决APT检测中的适应性不足问题 reinforcement learning PPO

🔬 支柱五：交互与反应 (Interaction & Reaction) (1 篇)

#题目一句话要点标签🔗⭐
40 Reflection-Enhanced Meta-Optimization Integrating TextGrad-style Prompt Optimization with Memory-Driven Self-Evolution 提出反射增强元优化框架以解决提示优化的历史经验利用问题 ReMoS large language model chain-of-thought

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

#题目一句话要点标签🔗⭐
41 Long-Term Variability in Physiological-Arousal Relationships for Robust Emotion Estimation 提出长期生理唤醒关系变异性研究以提升情感估计模型的鲁棒性 PULSE

⬅️ 返回 cs.AI 首页 · 🏠 返回主页