cs.AI（2025-04-30）

📊 共 13 篇论文 | 🔗 3 篇有代码

🎯 兴趣领域导航

支柱二：RL算法与架构 (RL & Architecture) (5) 支柱九：具身大模型 (Embodied Foundation Models) (5 🔗2) 支柱三：空间感知与语义 (Perception & Semantics) (1 🔗1) 支柱五：交互与反应 (Interaction & Reaction) (1) 支柱八：物理动画 (Physics-based Animation) (1)

🔬 支柱二：RL算法与架构 (RL & Architecture) (5 篇)

#题目一句话要点标签🔗⭐
1 Reinforced MLLM: A Survey on RL-Based Reasoning in Multimodal Large Language Models 综述：强化学习驱动多模态大语言模型推理能力提升 reinforcement learning large language model multimodal
2 Adaptive 3D UI Placement in Mixed Reality Using Deep Reinforcement Learning 提出基于深度强化学习的自适应3D UI放置方法，优化混合现实用户体验 reinforcement learning deep reinforcement learning
3 Designing Control Barrier Function via Probabilistic Enumeration for Safe Reinforcement Learning Navigation 提出基于概率枚举的控制屏障函数设计方法，用于安全强化学习导航 reinforcement learning
4 How to Backdoor the Knowledge Distillation 提出一种针对知识蒸馏的后门攻击方法，利用对抗样本毒化蒸馏数据集。 distillation
5 ShorterBetter: Guiding Reasoning Models to Find Optimal Inference Length for Efficient Reasoning ShorterBetter：引导推理模型学习最优推理长度，提升推理效率 reinforcement learning chain-of-thought

🔬 支柱九：具身大模型 (Embodied Foundation Models) (5 篇)

#题目一句话要点标签🔗⭐
6 Galvatron: An Automatic Distributed System for Efficient Foundation Model Training Galvatron：用于高效训练大模型自动分布式系统 foundation model ✅
7 MF-LLM: Simulating Population Decision Dynamics via a Mean-Field Large Language Model Framework 提出MF-LLM框架，通过均值场理论提升LLM在群体决策模拟中的真实性 large language model
8 IRL Dittos: Embodied Multimodal AI Agent Interactions in Open Spaces 提出IRL Ditto具身智能体，增强分布式团队在共享办公空间的社交互动。 multimodal
9 RAIL in the Wild: Operationalizing Responsible AI Evaluation Using Anthropic's Value Dataset 利用Anthropic价值观数据集，提出RAIL框架以评估LLM的伦理行为 large language model
10 Ada-R1: Hybrid-CoT via Bi-Level Adaptive Reasoning Optimization 提出Ada-R1，通过双层自适应推理优化实现高效混合CoT推理。 large language model ✅

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

#题目一句话要点标签🔗⭐
11 A Survey on 3D Reconstruction Techniques in Plant Phenotyping: From Classical Methods to Neural Radiance Fields (NeRF), 3D Gaussian Splatting (3DGS), and Beyond 综述植物表型3D重建技术：从传统方法到NeRF、3DGS及未来展望 3D gaussian splatting 3DGS gaussian splatting ✅

🔬 支柱五：交互与反应 (Interaction & Reaction) (1 篇)

#题目一句话要点标签🔗⭐
12 Efficient Quantum-Safe Homomorphic Encryption for Quantum Computer Programs 提出一种抗量子计算机攻击的高效同态加密方案，用于量子程序安全评估。 OMOMO

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

#题目一句话要点标签🔗⭐
13 Sionna RT: Technical Report Sionna RT：开源可微GPU加速射线追踪，用于无线信道仿真 PULSE

⬅️ 返回 cs.AI 首页 · 🏠 返回主页