cs.LG（2026-07-15）
📊 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 🔗1)
支柱二：RL算法与架构 (RL & Architecture) (5)
支柱一：机器人控制 (Robot Control) (2)
🔬 支柱九：具身大模型 (Embodied Foundation Models) (7 篇)
| # | 题目 | 一句话要点 | 标签 | 🔗 | ⭐ |
|---|---|---|---|---|---|
| 1 | MetaPerch: Learning from metadata for bioacoustics foundation models | 提出MetaPerch以利用元数据提升生物声学模型性能 | foundation model | ||
| 2 | Multimodal Empirical Bayes Variational Autoencoders for Joint Longitudinal and Time-to-Event Modeling | 提出多模态经验贝叶斯变分自编码器以解决肿瘤生长与事件时间建模问题 | multimodal | ||
| 3 | MxGPS: Multiplex Graph Transformers for a Power Grid Foundation Model | 提出MxGPS以解决电网模型的拓扑过拟合问题 | foundation model | ||
| 4 | PQFA: Parallel Quantum Feature Augmentation of Fused Representations for Multimodal Classification | 提出PQFA以增强多模态分类的后融合特征 | multimodal | ||
| 5 | Self-Improving is Often Sudden: Enlightenment-style Finetuning for Large-Scale Models | 提出启蒙式微调方法以提升大规模模型性能 | large language model foundation model | ||
| 6 | The Hyperspherical Geometry of CLIP Latent Space: A Semantic Mixture Model | 提出基于von Mises-Fisher混合模型的CLIP潜在空间密度估计方法 | multimodal | ✅ | |
| 7 | Data-Efficient Adaptation of LLMs via Attention Head Reweighting | 提出注意力头重标定方法以解决LLMs数据高效适应问题 | 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 | 🔗 | ⭐ |
|---|---|---|---|---|---|
| 8 | The SIGReg Objective as Variational Free Energy: A Theoretical Active-Inference Account of JEPA World Models | 提出SIGReg目标作为变分自由能以优化JEPA世界模型 | world model world models JEPA | ||
| 9 | Consensus as Privileged Context for Label-Free Self-Distillation | 提出CANON以解决无标签自蒸馏中的共识信息利用问题 | reinforcement learning distillation large language model | ||
| 10 | Structured Reinforcement Learning for Bayesian Persuasion : Application to Intelligent Interactive Driving | 提出结构化强化学习以解决贝叶斯劝说在智能驾驶中的应用问题 | reinforcement learning policy learning | ||
| 11 | Leveraging unlabelled data for generalizable neural population decoding | 提出MOJO框架以解决神经解码中的标签稀缺问题 | masked autoencoder foundation model | ||
| 12 | Where Should RL Post-Training Compute Go? Model Size, Search, Learning, and Feedback | 提出FLOP会计框架优化RL后训练资源分配问题 | reinforcement learning foundation model |
🔬 支柱一：机器人控制 (Robot Control) (2 篇)
| # | 题目 | 一句话要点 | 标签 | 🔗 | ⭐ |
|---|---|---|---|---|---|
| 13 | Factorized Spectral Representations for Reinforcement Learning | 提出FaStR以提升强化学习中的世界模型学习效率 | locomotion reinforcement learning deep reinforcement learning | ||
| 14 | DAGR: State-Conditioned Goal Representations via Difference-Aware Goal Cross-Attention | 提出DAGR以解决目标编码的状态依赖问题 | manipulation reinforcement learning |