cs.AI（2025-07-06）

📊 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) (7) 支柱二：RL算法与架构 (RL & Architecture) (5 🔗1) 支柱三：空间感知与语义 (Perception & Semantics) (1) 支柱一：机器人控制 (Robot Control) (1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (7 篇)

#题目一句话要点标签🔗⭐
1 The role of large language models in UI/UX design: A systematic literature review 系统性综述：大型语言模型在UI/UX设计中的应用与挑战 large language model multimodal
2 A validity-guided workflow for robust large language model research in psychology 提出基于效度引导的工作流，提升大语言模型在心理学研究中的可靠性。 large language model
3 LayerCake: Token-Aware Contrastive Decoding within Large Language Model Layers LayerCake：提出一种token感知的层内对比解码方法，提升大语言模型的事实性生成能力。 large language model
4 Gaze-Aware AI: Mathematical modeling of epistemic experience of the Marginalized for Human-Computer Interaction & AI Systems 提出Gaze Pressure Index (GPI)-Diff复合指标，用于量化边缘群体在人机交互中的认知体验。 large language model
5 Attention Slipping: A Mechanistic Understanding of Jailbreak Attacks and Defenses in LLMs 揭示LLM越狱攻击机制：注意力滑移现象与防御方法 large language model
6 Integrating Generative AI in BIM Education: Insights from Classroom Implementation 探索生成式AI在BIM教育中的应用：课堂实践与经验 large language model
7 Mpemba Effect in Large-Language Model Training Dynamics: A Minimal Analysis of the Valley-River model 基于Mpemba效应分析LLM训练动态，优化学习率Plateau策略 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🔗⭐
8 Anomalous Decision Discovery using Inverse Reinforcement Learning 提出基于逆强化学习的异常决策发现框架以解决自动驾驶中的异常检测问题 reinforcement learning inverse reinforcement learning ✅
9 CLIP-RL: Surgical Scene Segmentation Using Contrastive Language-Vision Pretraining & Reinforcement Learning 提出CLIP-RL，利用对比语言-视觉预训练和强化学习进行手术场景分割。 reinforcement learning contrastive learning curriculum learning
10 DC-Mamber: A Dual Channel Prediction Model based on Mamba and Linear Transformer for Multivariate Time Series Forecasting 提出基于Mamba和线性Transformer的双通道预测模型DC-Mamber，用于提升多元时间序列预测精度。 Mamba SSM state space model
11 WebSynthesis: World-Model-Guided MCTS for Efficient WebUI-Trajectory Synthesis WebSynthesis：利用世界模型引导的MCTS高效合成WebUI交互轨迹 world model large language model
12 VOLTRON: Detecting Unknown Malware Using Graph-Based Zero-Shot Learning 提出基于图的零样本学习框架VOLTRON，用于检测未知恶意软件。 Voltron

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

#题目一句话要点标签🔗⭐
13 High-Resolution Sustain Pedal Depth Estimation from Piano Audio Across Room Acoustics 提出基于Transformer的钢琴延音踏板深度估计方法，提升音乐表现力并分析房间声学影响。 depth estimation

🔬 支柱一：机器人控制 (Robot Control) (1 篇)

#题目一句话要点标签🔗⭐
14 Hijacking JARVIS: Benchmarking Mobile GUI Agents against Unprivileged Third Parties AgentHazard：首个针对移动GUI代理抗恶意第三方内容攻击的评测基准 manipulation vision-language-action

⬅️ 返回 cs.AI 首页 · 🏠 返回主页