cs.AI（2026-01-12）

📊 共 31 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) (18 🔗2) 支柱二：RL算法与架构 (RL & Architecture) (10) 支柱一：机器人控制 (Robot Control) (1) 支柱六：视频提取与匹配 (Video Extraction) (1) 支柱八：物理动画 (Physics-based Animation) (1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (18 篇)

#题目一句话要点标签🔗⭐
1 Learning to Trust the Crowd: A Multi-Model Consensus Reasoning Engine for Large Language Models 提出多模型共识推理引擎，提升大语言模型在实例层面的可靠性。 large language model
2 Safe-FedLLM: Delving into the Safety of Federated Large Language Models Safe-FedLLM：提出一种基于探针的联邦LLM防御框架，提升对抗恶意客户端攻击的安全性。 large language model ✅
3 IFDNS: An Iterative Feedback-Driven Neuro-Symbolic Method for Faithful Logical Reasoning 提出IFDNS：一种迭代反馈驱动的神经符号方法，用于可信的逻辑推理 large language model chain-of-thought
4 OS-Symphony: A Holistic Framework for Robust and Generalist Computer-Using Agent OS-Symphony：用于提升计算机使用Agent鲁棒性和泛化性的整体框架 multimodal
5 ARM: Role-Conditioned Neuron Transplantation for Training-Free Generalist LLM Agent Merging 提出ARM：一种免训练的角色条件神经元移植方法，用于通用LLM Agent融合。 large language model
6 SALT-KG: A Benchmark for Semantics-Aware Learning on Enterprise Tables SALT-KG：一个用于企业表格语义感知学习的基准数据集 foundation model
7 Beyond Entangled Planning: Task-Decoupled Planning for Long-Horizon Agents 提出任务解耦规划（TDP）框架，提升长程Agent任务执行的鲁棒性和效率 large language model
8 Beyond Dialogue Time: Temporal Semantic Memory for Personalized LLM Agents 提出时间语义记忆（TSM）框架，解决LLM Agent中记忆的时间不准确和碎片化问题 large language model
9 RLPO: Residual Listwise Preference Optimization for Long-Context Review Ranking 提出RLPO：一种残差列表偏好优化方法，用于长文本评论排序。 large language model
10 Agentic Diagnostic Reasoning over Telecom and Datacenter Infrastructure 提出基于LLM Agent的诊断框架，用于电信和数据中心基础设施的故障根因分析。 large language model
11 When Bots Take the Bait: Exposing and Mitigating the Emerging Social Engineering Attack in Web Automation Agent 提出AgentBait攻击与SUPERVISOR防御，提升Web自动化Agent安全性 large language model
12 Stochastic CHAOS: Why Deterministic Inference Kills, and Distributional Variability Is the Heartbeat of Artifical Cognition 反对LLM确定性推理：提出Stochastic CHAOS以提升模型不确定性建模与安全性 large language model
13 From "Thinking" to "Justifying": Aligning High-Stakes Explainability with Professional Communication Standards 提出结构化解释框架以提升高风险领域的可解释性 chain-of-thought
14 DiSCo: Making Absence Visible in Intelligent Summarization Interfaces DiSCo通过对比领域知识，使智能摘要界面中信息的缺失变得可见。 large language model
15 LLMRouterBench: A Massive Benchmark and Unified Framework for LLM Routing LLMRouterBench：大规模LLM路由基准测试与统一框架 large language model ✅
16 Active Context Compression: Autonomous Memory Management in LLM Agents Focus：面向LLM Agent的主动上下文压缩，解决长程任务中的Context Bloat问题 large language model
17 Defenses Against Prompt Attacks Learn Surface Heuristics 提出对抗提示攻击的新防御方法以解决现有模型的安全性问题 large language model
18 A Large-Scale Study on the Development and Issues of Multi-Agent AI Systems 大规模分析多智能体AI系统演进与问题，揭示开发挑战与维护需求。 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🔗⭐
19 Large Language Models for Physics Instrument Design 利用大型语言模型进行物理仪器设计，探索LLM在复杂科学问题中的应用潜力 reinforcement learning large language model
20 Puzzle it Out: Local-to-Global World Model for Offline Multi-Agent Reinforcement Learning 提出局部到全局世界模型LOGO，解决离线多智能体强化学习泛化性问题 reinforcement learning policy learning world model
21 ENTRA: Entropy-Based Redundancy Avoidance in Large Language Model Reasoning ENTRA：提出基于熵的冗余避免框架，提升大语言模型推理效率。 reinforcement learning large language model
22 Rewarding Creativity: A Human-Aligned Generative Reward Model for Reinforcement Learning in Storytelling 提出人类对齐的生成奖励模型以解决创意故事生成中的挑战 reinforcement learning reward shaping large language model
23 AscendKernelGen: A Systematic Study of LLM-Based Kernel Generation for Neural Processing Units AscendKernelGen：面向昇腾NPU的LLM内核生成系统性研究 reinforcement learning large language model chain-of-thought
24 Enhancing Cloud Network Resilience via a Robust LLM-Empowered Multi-Agent Reinforcement Learning Framework 提出CyberOps-Bots，一种基于LLM的多智能体强化学习框架，提升云网络弹性。 reinforcement learning large language model
25 Knowledge Distillation for LLM-Based Human Activity Recognition in Homes 利用知识蒸馏提升LLM在家庭环境人体活动识别中的效率 distillation large language model
26 OpenTinker: Separating Concerns in Agentic Reinforcement Learning OpenTinker：面向Agent强化学习的关注点分离基础设施 reinforcement learning large language model
27 Learning How to Remember: A Meta-Cognitive Management Method for Structured and Transferable Agent Memory 提出MCMA方法以解决大语言模型记忆管理问题 direct preference optimization large language model
28 LRAS: Advanced Legal Reasoning with Agentic Search 提出LRAS框架，通过Agentic Search提升法律大语言模型推理能力。 reinforcement learning imitation learning

🔬 支柱一：机器人控制 (Robot Control) (1 篇)

#题目一句话要点标签🔗⭐
29 VirtualEnv: A Platform for Embodied AI Research VirtualEnv：一个用于具身人工智能研究的交互式模拟平台 manipulation embodied AI large language model

🔬 支柱六：视频提取与匹配 (Video Extraction) (1 篇)

#题目一句话要点标签🔗⭐
30 Yes FLoReNce, I Will Do Better Next Time! Agentic Feedback Reasoning for Humorous Meme Detection 提出FLoReNce框架，通过Agent反馈推理提升幽默Meme检测性能 HuMoR multimodal

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

#题目一句话要点标签🔗⭐
31 Efficient Convolutional Forward Model for Passive Acoustic Mapping and Temporal Monitoring 提出一种高效卷积前向模型，用于被动声学成像和时间监测 spatiotemporal

⬅️ 返回 cs.AI 首页 · 🏠 返回主页