cs.LG（2025-12-27）

📊 共 11 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) (4 🔗1) 支柱一：机器人控制 (Robot Control) (3) 支柱二：RL算法与架构 (RL & Architecture) (2) 支柱七：动作重定向 (Motion Retargeting) (1) 支柱四：生成式动作 (Generative Motion) (1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (4 篇)

#题目一句话要点标签🔗⭐
1 Scaling Unverifiable Rewards: A Case Study on Visual Insights 提出Selective TTS，通过阶段性推理优化解决不可验证奖励下的多阶段任务。 large language model
2 AdaFRUGAL: Adaptive Memory-Efficient Training with Dynamic Control AdaFRUGAL：通过动态控制实现自适应、内存高效的大模型训练 large language model
3 TimePerceiver: An Encoder-Decoder Framework for Generalized Time-Series Forecasting TimePerceiver：一种用于广义时间序列预测的Encoder-Decoder框架 TAMP ✅
4 The Bayesian Geometry of Transformer Attention Transformer通过几何机制实现贝叶斯推理，显著优于MLP large language model

🔬 支柱一：机器人控制 (Robot Control) (3 篇)

#题目一句话要点标签🔗⭐
5 Communication Compression for Distributed Learning with Aggregate and Server-Guided Feedback 提出CAFe和CAFe-S框架，解决联邦学习中通信压缩和隐私保护的难题。 shared control
6 Decomposing Task Vectors for Refined Model Editing 提出任务向量分解方法，实现对预训练模型行为的精确控制与编辑 manipulation
7 Geometric Scaling of Bayesian Inference in LLMs 探讨大语言模型中贝叶斯推断的几何特征 manipulation

🔬 支柱二：RL算法与架构 (RL & Architecture) (2 篇)

#题目一句话要点标签🔗⭐
8 Toward Real-World IoT Security: Concept Drift-Resilient IoT Botnet Detection via Latent Space Representation Learning and Alignment 提出基于潜在空间对齐的IoT僵尸网络检测框架，解决概念漂移问题 representation learning
9 Predictive Modeling of Power Outages during Extreme Events: Integrating Weather and Socio-Economic Factors 提出一种融合天气与社会经济因素的预测模型，用于极端事件下的电力中断预测。 predictive model

🔬 支柱七：动作重定向 (Motion Retargeting) (1 篇)

#题目一句话要点标签🔗⭐
10 Uncertainty-Aware Flow Field Reconstruction Using SVGP Kolmogorov-Arnold Networks 提出基于SVGP-KAN的不确定性感知流场重建方法，解决时序稀疏数据下的流场重建问题。 structure preservation PULSE

🔬 支柱四：生成式动作 (Generative Motion) (1 篇)

#题目一句话要点标签🔗⭐
11 Quantum Generative Models for Computational Fluid Dynamics: A First Exploration of Latent Space Learning in Lattice Boltzmann Simulations 首次探索量子生成模型在CFD中的应用，用于学习LBM模拟的潜在空间表示 VQ-VAE

⬅️ 返回 cs.LG 首页 · 🏠 返回主页