cs.AI（2025-04-17）

📊 共 16 篇论文 | 🔗 3 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (9 🔗2) 支柱二：RL算法与架构 (RL & Architecture) (6 🔗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 Towards Cardiac MRI Foundation Models: Comprehensive Visual-Tabular Representations for Whole-Heart Assessment and Beyond ViTa：面向心脏MRI的Foundation Model，融合视觉-表格数据实现全面心脏评估 foundation model
2 The Future of Internet of Things and Multimodal Language Models in 6G Networks: Opportunities and Challenges 综述：6G网络中物联网与多模态语言模型融合的机遇与挑战 multimodal
3 EmoVoice: LLM-based Emotional Text-To-Speech Model with Freestyle Text Prompting EmoVoice：基于LLM和自由文本提示的情感可控语音合成模型 large language model multimodal chain-of-thought ✅
4 SimUSER: Simulating User Behavior with Large Language Models for Recommender System Evaluation SimUSER：利用大语言模型模拟用户行为，用于推荐系统评估 large language model
5 Causal-Copilot: An Autonomous Causal Analysis Agent Causal-Copilot：基于大语言模型的自主因果分析Agent，赋能领域专家。 large language model
6 Sleep-time Compute: Beyond Inference Scaling at Test-time 提出睡眠时间计算，通过离线预计算减少LLM推理时延与成本。 large language model
7 Exploring Expert Failures Improves LLM Agent Tuning EEF：利用专家失败经验提升LLM Agent的微调效果，刷新WebShop和SciWorld记录。 large language model
8 ZeroSumEval: Scaling LLM Evaluation with Inter-Model Competition ZeroSumEval：利用模型间零和博弈扩展LLM评估，解决传统评估方法的局限性。 large language model ✅
9 Knowledge Acquisition on Mass-shooting Events via LLMs for AI-Driven Justice 利用大型语言模型进行大规模枪击事件知识获取，助力AI驱动的司法 large language model

🔬 支柱二：RL算法与架构 (RL & Architecture) (6 篇)

#题目一句话要点标签🔗⭐
10 Embodied-R: Collaborative Framework for Activating Embodied Spatial Reasoning in Foundation Models via Reinforcement Learning Embodied-R：通过强化学习激活具身空间推理能力的基础模型协同框架 reinforcement learning reward design egocentric
11 Governance Challenges in Reinforcement Learning from Human Feedback: Evaluator Rationality and Reinforcement Stability 研究表明评估者理性程度影响RLHF稳定性，并提出改进RLHF治理的建议 reinforcement learning RLHF large language model
12 QLLM: Do We Really Need a Mixing Network for Credit Assignment in Multi-Agent Reinforcement Learning? QLLM：利用大语言模型自动构建信用分配函数，提升多智能体强化学习性能。 reinforcement learning large language model ✅
13 Antidistillation Sampling 提出Antidistillation Sampling，通过毒化推理轨迹提升模型蒸馏防御能力。 distillation
14 InstructRAG: Leveraging Retrieval-Augmented Generation on Instruction Graphs for LLM-Based Task Planning 提出InstructRAG，利用指令图上的检索增强生成提升LLM在任务规划中的性能。 reinforcement learning large language model
15 Multi-Agent Reinforcement Learning Simulation for Environmental Policy Synthesis 提出基于多智能体强化学习的气候政策合成框架，应对气候政策制定的挑战。 reinforcement learning

🔬 支柱一：机器人控制 (Robot Control) (1 篇)

#题目一句话要点标签🔗⭐
16 Security-First AI: Foundations for Robust and Trustworthy Systems 提出“安全优先”AI框架，保障AI系统鲁棒性和可信性 manipulation

⬅️ 返回 cs.AI 首页 · 🏠 返回主页