cs.AI（2026-02-17）

📊 共 14 篇论文 | 🔗 1 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (9 🔗1) 支柱二：RL算法与架构 (RL & Architecture) (3) 支柱八：物理动画 (Physics-based Animation) (1) 支柱一：机器人控制 (Robot Control) (1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (9 篇)

#题目一句话要点标签🔗⭐
1 Quantifying construct validity in large language model evaluations 提出结构化能力模型，用于量化大语言模型评估中的构建效度 large language model
2 Enhancing Building Semantics Preservation in AI Model Training with Large Language Model Encodings 利用大语言模型编码增强AI模型训练中建筑语义的保持能力 large language model
3 How Vision Becomes Language: A Layer-wise Information-Theoretic Analysis of Multimodal Reasoning 提出基于信息论的层级分析框架，解析多模态Transformer的推理机制。 multimodal
4 CARE Drive A Framework for Evaluating Reason-Responsiveness of Vision Language Models in Automated Driving CARE Drive：评估自动驾驶中视觉语言模型对人类理由的响应性 foundation model
5 Decision Quality Evaluation Framework at Pinterest Pinterest提出决策质量评估框架，用于提升内容安全策略执行效果。 large language model
6 This human study did not involve human subjects: Validating LLM simulations as behavioral evidence 验证LLM模拟作为行为证据：探索LLM在社会科学实验中的有效性 large language model
7 SecCodeBench-V2 Technical Report SecCodeBench-V2：一个用于评估LLM代码生成安全性的工业级基准 large language model ✅
8 Automated Multi-Source Debugging and Natural Language Error Explanation for Dashboard Applications 提出一种自动多源调试与自然语言错误解释框架，用于改进仪表盘应用的调试效率。 large language model
9 EAA: Automating materials characterization with vision language model agents EAA：利用视觉语言模型智能体自动化材料表征实验流程 multimodal

🔬 支柱二：RL算法与架构 (RL & Architecture) (3 篇)

#题目一句话要点标签🔗⭐
10 Recursive Concept Evolution for Compositional Reasoning in Large Language Models 提出递归概念演化（RCE）框架，提升大语言模型在组合推理任务上的性能。 reinforcement learning large language model chain-of-thought
11 World-Model-Augmented Web Agents with Action Correction 提出WAC，通过世界模型增强Web Agent，并进行动作校正，提升Web任务自动化能力。 world model large language model
12 SCENE OTA-FD: Self-Centering Noncoherent Estimator for Over-the-Air Federated Distillation 提出SCENE，用于无线联邦蒸馏的自定心非相干估计器 distillation

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

#题目一句话要点标签🔗⭐
13 AgriWorld:A World Tools Protocol Framework for Verifiable Agricultural Reasoning with Code-Executing LLM Agents 提出AgriWorld框架以解决农业推理与交互能力不足的问题 spatiotemporal large language model foundation model

🔬 支柱一：机器人控制 (Robot Control) (1 篇)

#题目一句话要点标签🔗⭐
14 PERSONA: Dynamic and Compositional Inference-Time Personality Control via Activation Vector Algebra PERSONA：通过激活向量代数实现大语言模型中动态和可组合的推理时人格控制 manipulation large language model

⬅️ 返回 cs.AI 首页 · 🏠 返回主页