cs.AI（2025-09-29）

📊 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) (24 🔗2) 支柱二：RL算法与架构 (RL & Architecture) (13) 支柱三：空间感知与语义 (Perception & Semantics) (1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (24 篇)

#题目一句话要点标签🔗⭐
1 Building the EHR Foundation Model via Next Event Prediction 提出基于事件预测的EHR基础模型，增强LLM在临床时序推理能力 large language model foundation model
2 Radiology's Last Exam (RadLE): Benchmarking Frontier Multimodal AI Against Human Experts and a Taxonomy of Visual Reasoning Errors in Radiology RadLE：放射学诊断基准测试，评估多模态AI与专家医生的差距及视觉推理错误 large language model multimodal
3 TimeOmni-1: Incentivizing Complex Reasoning with Time Series in Large Language Models TimeOmni-1：通过时间序列激励大语言模型进行复杂推理 large language model multimodal
4 Bridging the behavior-neural gap: A multimodal AI reveals the brain's geometry of emotion more accurately than human self-reports 多模态AI超越人类自报告，更准确揭示大脑情感几何 large language model multimodal ✅
5 Model Merging Scaling Laws in Large Language Models 提出语言模型融合的规模法则，实现专家模型高效组合与性能预测 large language model
6 Chat to Chip: Large Language Model Based Design of Arbitrarily Shaped Metasurfaces 提出基于大语言模型的超表面设计方法，实现任意形状超表面的光谱预测与逆向设计。 large language model
7 Evaluating Foundation Models with Pathological Concept Learning for Kidney Cancer 提出基于病理概念学习的肾癌评估方法，利用基础模型提升生存分析效果。 foundation model ✅
8 AdvChain: Adversarial Chain-of-Thought Tuning for Robust Safety Alignment of Large Reasoning Models AdvChain：对抗式思维链调优，提升大型推理模型安全对齐的鲁棒性 chain-of-thought
9 ELHPlan: Efficient Long-Horizon Task Planning for Multi-Agent Collaboration ELHPlan：面向多智能体协作的高效长时程任务规划框架 large language model
10 Advancing mathematics research with generative AI 利用生成式AI辅助数学研究，提升问题求解与猜想能力 large language model
11 TENET: Leveraging Tests Beyond Validation for Code Generation TENET：利用测试驱动开发提升代码生成质量，解决复杂仓库环境下的代码生成难题。 large language model
12 MGM-Omni: Scaling Omni LLMs to Personalized Long-Horizon Speech MGM-Omni：面向个性化长时程语音的通用多模态大语言模型 multimodal
13 Causal Autoencoder-like Generation of Feedback Fuzzy Cognitive Maps with an LLM Agent 提出基于LLM的因果模糊认知图自编码器，实现可解释的认知图重建。 large language model
14 ATLAS: Constraints-Aware Multi-Agent Collaboration for Real-World Travel Planning ATLAS：面向真实旅行规划的约束感知多智能体协作框架 large language model
15 A(I)nimism: Re-enchanting the World Through AI-Mediated Object Interaction 提出A(I)nimism以重塑人与物的互动关系 large language model
16 Toxicity in Online Platforms and AI Systems: A Survey of Needs, Challenges, Mitigations, and Future Directions 构建在线平台和AI系统毒性全面分类体系，旨在促进毒性检测与缓解方案设计。 large language model
17 Adaptive Test-Time Reasoning via Reward-Guided Dual-Phase Search 提出基于奖励引导的双阶段搜索，提升LLM在推理任务中的效率和准确性。 large language model
18 AutoCode: LLMs as Problem Setters for Competitive Programming AutoCode：利用大语言模型自动生成高质量的竞赛编程题目 large language model
19 ReasoningBank: Scaling Agent Self-Evolving with Reasoning Memory 提出 ReasoningBank，通过推理记忆和自进化提升Agent在持续任务中的性能。 large language model
20 Dive into the Agent Matrix: A Realistic Evaluation of Self-Replication Risk in LLM Agents 构建真实场景评估LLM Agent的自复制风险，揭示潜在安全隐患 large language model
21 Flash-Searcher: Fast and Effective Web Agents via DAG-Based Parallel Execution Flash-Searcher：基于DAG并行执行的快速高效Web Agent large language model
22 MASLegalBench: Benchmarking Multi-Agent Systems in Deductive Legal Reasoning 提出MASLegalBench：用于评估多智能体系统在演绎法律推理中的性能 large language model
23 Neural network embeddings recover value dimensions from psychometric survey items on par with human data 利用神经网络嵌入和SQuID方法，从心理测量问卷条目中恢复人类价值观维度，效果与人类数据相当 large language model
24 Experience-Guided Reflective Co-Evolution of Prompts and Heuristics for Automatic Algorithm Design 提出EvoPH框架，通过经验引导的提示与启发式算法协同进化，实现自动算法设计 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🔗⭐
25 Uni-NTFM: A Unified Foundation Model for EEG Signal Representation Learning Uni-NTFM：用于脑电信号表征学习的统一神经拓扑基础模型 representation learning foundation model
26 RE-PO: Robust Enhanced Policy Optimization as a General Framework for LLM Alignment RE-PO：一种通用的LLM对齐框架，通过鲁棒增强策略优化解决标签噪声问题 reinforcement learning RLHF DPO
27 Training Agents Inside of Scalable World Models Dreamer 4：通过可扩展世界模型在Minecraft中实现离线钻石获取 reinforcement learning world model dreamer
28 Reasoning or Retrieval? A Study of Answer Attribution on Large Reasoning Models 揭示大语言模型推理与检索的竞争机制，提出FARL提升推理能力 reinforcement learning distillation chain-of-thought
29 RL in the Wild: Characterizing RLVR Training in LLM Deployment 针对LLM部署中RLVR训练的系统挑战，提出PolyTrace基准测试套件。 reinforcement learning large language model
30 Hybrid Reward Normalization for Process-supervised Non-verifiable Agentic Tasks 提出原则过程奖励以解决长轨迹任务中的反馈稀疏问题 reinforcement learning large language model
31 Modeling Others' Minds as Code ROTE：利用程序合成高效预测人类行为，提升人机协作 behavior cloning large language model
32 DeepSearch: Overcome the Bottleneck of Reinforcement Learning with Verifiable Rewards via Monte Carlo Tree Search DeepSearch：通过蒙特卡洛树搜索和可验证奖励克服强化学习瓶颈 reinforcement learning
33 The Era of Real-World Human Interaction: RL from User Conversations 提出基于用户对话的强化学习(RLHI)，实现持续模型改进和多方面对齐。 reinforcement learning instruction following
34 Pushing LLMs to Their Logical Reasoning Bound: The Role of Data Reasoning Intensity 提出数据推理强度（DRI）指标，优化训练数据以提升LLM逻辑推理能力。 reinforcement learning large language model
35 Towards Safe Reasoning in Large Reasoning Models via Corrective Intervention 提出Intervened Preference Optimization以提升大型推理模型安全性 preference learning chain-of-thought
36 Humanline: Online Alignment as Perceptual Loss 提出Humanline，通过感知损失在线对齐，提升模型与人类偏好一致性 PPO DPO
37 Unifying Agent Interaction and World Information for Multi-agent Coordination 提出IWoL框架，统一交互与世界信息，促进多智能体协同 reinforcement learning representation learning

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

#题目一句话要点标签🔗⭐
38 Vision-and-Language Navigation with Analogical Textual Descriptions in LLMs 提出基于LLM和类比文本描述的视觉-语言导航方法，提升场景理解和空间推理能力 scene understanding embodied AI VLN

⬅️ 返回 cs.AI 首页 · 🏠 返回主页