cs.AI（2025-12-11）

📊 共 18 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) (11 🔗1) 支柱二：RL算法与架构 (RL & Architecture) (5 🔗1) 支柱四：生成式动作 (Generative Motion) (1) 支柱一：机器人控制 (Robot Control) (1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (11 篇)

#题目一句话要点标签🔗⭐
1 KathDB: Explainable Multimodal Database Management System with Human-AI Collaboration 提出KathDB，一个支持人机协作的可解释多模态数据库管理系统 foundation model multimodal
2 PACIFIC: a framework for generating benchmarks to check Precise Automatically Checked Instruction Following In Code PACIFIC：用于生成基准以评估LLM代码指令遵循能力的框架 large language model instruction following
3 A probabilistic foundation model for crystal structure denoising, phase classification, and order parameters 提出一种概率基础模型以解决晶体结构去噪与相分类问题 foundation model
4 Developing and Evaluating a Large Language Model-Based Automated Feedback System Grounded in Evidence-Centered Design for Supporting Physics Problem Solving 提出基于证据中心设计的LLM物理问题自动反馈系统，但需警惕其潜在错误。 large language model
5 Offscript: Automated Auditing of Instruction Adherence in LLMs Offscript：自动化审计LLM指令遵循情况，发现潜在违规行为。 large language model instruction following
6 Multi-Granular Node Pruning for Circuit Discovery 提出多粒度节点剪枝方法，用于大规模语言模型中的电路发现，提升效率和精度。 large language model
7 LLMs Can Assist with Proposal Selection at Large User Facilities 利用大型语言模型优化大型用户设施的提案选择过程 large language model
8 Natural Language Interface for Firewall Configuration 提出一种自然语言防火墙配置接口，简化企业网络安全策略管理。 large language model
9 CXL-SpecKV: A Disaggregated FPGA Speculative KV-Cache for Datacenter LLM Serving 提出CXL-SpecKV，利用CXL互连和FPGA加速实现LLM推理中KV缓存的卸载与加速。 large language model ✅
10 Challenges of Evaluating LLM Safety for User Welfare 针对用户福祉，提出考虑用户情境的LLM安全评估方法 large language model
11 On the Dynamics of Multi-Agent LLM Communities Driven by Value Diversity 探索价值观多样性对多智能体LLM社区集体行为的影响 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🔗⭐
12 Achieving Olympia-Level Geometry Large Language Model Agent via Complexity Boosting Reinforcement Learning 提出InternGeometry，通过复杂度提升强化学习实现奥赛级几何问题求解能力 reinforcement learning IMoS large language model
13 Enhancing Radiology Report Generation and Visual Grounding using Reinforcement Learning 利用强化学习增强放射报告生成和视觉定位，提升医学VLM性能 reinforcement learning visual grounding
14 Refinement Contrastive Learning of Cell-Gene Associations for Unsupervised Cell Type Identification 提出scRCL框架，通过细化对比学习细胞-基因关联，实现无监督细胞类型识别。 representation learning contrastive learning ✅
15 CORL: Reinforcement Learning of MILP Policies Solved via Branch and Bound 提出CORL框架，通过强化学习端到端优化分支定界求解的MILP策略 reinforcement learning
16 Remember Me, Refine Me: A Dynamic Procedural Memory Framework for Experience-Driven Agent Evolution ReMe：提出一种动态程序记忆框架，驱动Agent经验式进化 distillation large language model

🔬 支柱四：生成式动作 (Generative Motion) (1 篇)

#题目一句话要点标签🔗⭐
17 Phythesis: Physics-Guided Evolutionary Scene Synthesis for Energy-Efficient Data Center Design via LLMs Phythesis：利用LLM和物理引导进化算法实现节能数据中心场景合成 physically plausible large language model

🔬 支柱一：机器人控制 (Robot Control) (1 篇)

#题目一句话要点标签🔗⭐
18 When Reject Turns into Accept: Quantifying the Vulnerability of LLM-Based Scientific Reviewers to Indirect Prompt Injection 量化LLM科学评审员对间接Prompt注入的脆弱性，揭示拒稿变录用的风险 manipulation

⬅️ 返回 cs.AI 首页 · 🏠 返回主页