cs.AI（2025-12-17）

📊 共 16 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) (10 🔗1) 支柱二：RL算法与架构 (RL & Architecture) (5) 支柱一：机器人控制 (Robot Control) (1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (10 篇)

#题目一句话要点标签🔗⭐
1 Evaluating Large Language Models in Scientific Discovery 提出科学发现评估框架SDE，用于评估大语言模型在科学研究中的能力 large language model
2 Explaining the Reasoning of Large Language Models Using Attribution Graphs 提出CAGE框架，利用归因图解释大型语言模型推理过程，提升归因忠实度。 large language model
3 Beyond Accuracy: A Geometric Stability Analysis of Large Language Models in Chess Evaluation 提出几何稳定性框架，评估大语言模型在棋类评估中的推理能力，揭示准确率与鲁棒性悖论。 large language model
4 ChatGPT and Gemini participated in the Korean College Scholastic Ability Test -- Earth Science I 分析大型语言模型在韩国高考地球科学I科目的表现，揭示其认知局限性，为设计抗AI考题提供依据。 large language model multimodal
5 How Do Semantically Equivalent Code Transformations Impact Membership Inference on LLMs for Code? 语义等价代码变换削弱代码大语言模型成员推理攻击的有效性 large language model
6 On Assessing the Relevance of Code Reviews Authored by Generative Models 提出多主观排序评估方法，评估生成模型在代码评审中的有效性 large language model
7 Bilateral Spatial Reasoning about Street Networks: Graph-based RAG with Qualitative Spatial Representations 提出基于图的RAG方法，利用定性空间关系增强LLM的街道网络路径引导能力 large language model
8 SCOPE: Prompt Evolution for Enhancing Agent Effectiveness SCOPE：通过提示进化增强Agent有效性，解决大规模动态上下文中的管理瓶颈。 large language model ✅
9 Exploring User Acceptance and Concerns toward LLM-powered Conversational Agents in Immersive Extended Reality 探索用户对XR沉浸式环境中LLM驱动对话代理的接受度与担忧 large language model
10 Offline Multi-Task Multi-Objective Data-Driven Evolutionary Algorithm with Language Surrogate Model and Implicit Q-Learning 提出Q-MetaSur，利用语言模型和强化学习解决离线多任务多目标优化问题 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🔗⭐
11 Beyond Fast and Slow: Cognitive-Inspired Elastic Reasoning for Large Language Models 提出CogER框架，通过认知启发的弹性推理提升大语言模型在不同难度问题上的效率与准确性。 reinforcement learning large language model chain-of-thought
12 LADY: Linear Attention for Autonomous Driving Efficiency without Transformers 提出LADY：一种基于线性注意力的高效自动驾驶模型，无需Transformer。 linear attention spatiotemporal
13 Stepwise Think-Critique: A Unified Framework for Robust and Interpretable LLM Reasoning 提出Stepwise Think-Critique框架，提升LLM推理能力和可解释性 reinforcement learning large language model
14 Nemotron-Math: Efficient Long-Context Distillation of Mathematical Reasoning from Multi-Mode Supervision Nemotron-Math：通过多模式监督，高效地进行数学推理长文本蒸馏。 distillation
15 Graph Contextual Reinforcement Learning for Efficient Directed Controller Synthesis 提出GCRL，利用图上下文强化学习高效合成有向控制器 reinforcement learning

🔬 支柱一：机器人控制 (Robot Control) (1 篇)

#题目一句话要点标签🔗⭐
16 QoS-Aware Hierarchical Reinforcement Learning for Joint Link Selection and Trajectory Optimization in SAGIN-Supported UAV Mobility Management 提出基于QoS感知的分层强化学习方法，解决SAGIN支持的UAV移动性管理中的联合链路选择和轨迹优化问题。 trajectory optimization reinforcement learning deep reinforcement learning

⬅️ 返回 cs.AI 首页 · 🏠 返回主页