cs.LG(2024-05-21)
📊 共 8 篇论文
🎯 兴趣领域导航
支柱九:具身大模型 (Embodied Foundation Models) (3)
支柱一:机器人控制 (Robot Control) (2)
支柱二:RL算法与架构 (RL & Architecture) (2)
支柱八:物理动画 (Physics-based Animation) (1)
🔬 支柱九:具身大模型 (Embodied Foundation Models) (3 篇)
| # | 题目 | 一句话要点 | 标签 | 🔗 | ⭐ |
|---|---|---|---|---|---|
| 1 | Reducing Transformer Key-Value Cache Size with Cross-Layer Attention | 提出跨层注意力(CLA)机制,在保证精度前提下进一步压缩Transformer KV缓存 | large language model | ||
| 2 | SPO: Multi-Dimensional Preference Sequential Alignment With Implicit Reward Modeling | 提出SPO方法以解决人类偏好多维对齐问题 | large language model | ||
| 3 | TrajCogn: Leveraging LLMs for Cognizing Movement Patterns and Travel Purposes from Trajectories | TrajCogn:利用LLM认知轨迹中的移动模式和出行目的 | large language model |
🔬 支柱一:机器人控制 (Robot Control) (2 篇)
| # | 题目 | 一句话要点 | 标签 | 🔗 | ⭐ |
|---|---|---|---|---|---|
| 4 | Efficient Imitation Learning with Conservative World Models | 提出基于保守世界模型的模仿学习方法,提升样本效率和泛化能力 | manipulation reinforcement learning policy learning | ||
| 5 | Transformer in Touch: A Survey | 综述Transformer在触觉感知领域的应用,探索其在对象识别、跨模态生成和对象操作中的潜力。 | manipulation |
🔬 支柱二:RL算法与架构 (RL & Architecture) (2 篇)
| # | 题目 | 一句话要点 | 标签 | 🔗 | ⭐ |
|---|---|---|---|---|---|
| 6 | Aligning Transformers with Continuous Feedback via Energy Rank Alignment | 提出能量排序对齐(ERA)算法,用于优化自回归策略以生成具有指定属性的分子和蛋白质序列。 | reinforcement learning PPO DPO | ||
| 7 | GASE: Graph Attention Sampling with Edges Fusion for Solving Vehicle Routing Problems | 提出GASE框架,通过图注意力采样与边融合解决车辆路径问题 | reinforcement learning deep reinforcement learning |
🔬 支柱八:物理动画 (Physics-based Animation) (1 篇)
| # | 题目 | 一句话要点 | 标签 | 🔗 | ⭐ |
|---|---|---|---|---|---|
| 8 | A finite element-based physics-informed operator learning framework for spatiotemporal partial differential equations on arbitrary domains | 提出基于有限元的物理信息算子学习框架,用于预测任意域上的时空偏微分方程 | spatiotemporal |