| 1 |
QE-Catalytic: A Graph-Language Multimodal Base Model for Relaxed-Energy Prediction in Catalytic Adsorption |
提出QE-Catalytic，融合图和语言模型，提升催化吸附中弛豫能量预测精度。 |
MAE large language model multimodal |
|
|
| 2 |
Sample-Efficient Policy Constraint Offline Deep Reinforcement Learning based on Sample Filtering |
提出基于样本过滤的策略约束离线深度强化学习方法，提升样本效率。 |
reinforcement learning deep reinforcement learning offline RL |
|
|
| 3 |
Recurrent Off-Policy Deep Reinforcement Learning Doesn't Have to be Slow |
提出RISE，通过简化编码提升图像Off-Policy强化学习中循环网络的效率 |
reinforcement learning deep reinforcement learning |
|
|
| 4 |
TableGPT-R1: Advancing Tabular Reasoning Through Reinforcement Learning |
TableGPT-R1：通过强化学习提升表格推理能力，实现SOTA性能。 |
reinforcement learning reward shaping large language model |
✅ |
|
| 5 |
Emergent temporal abstractions in autoregressive models enable hierarchical reinforcement learning |
提出内部强化学习，利用自回归模型中的时间抽象实现分层强化学习 |
reinforcement learning foundation model |
|
|
| 6 |
Performative Policy Gradient: Optimality in Performative Reinforcement Learning |
提出PePG算法，解决强化学习中策略执行带来的环境动态变化问题，实现策略的执行最优性。 |
reinforcement learning |
|
|
| 7 |
Jensen-Shannon Divergence Message-Passing for Rich-Text Graph Representation Learning |
提出JSDMP框架，利用Jensen-Shannon散度提升富文本图表示学习 |
representation learning |
|
|
| 8 |
Generalisation in Multitask Fitted Q-Iteration and Offline Q-learning |
提出多任务离线Q学习方法以提升统计效率与泛化能力 |
reinforcement learning offline reinforcement learning |
|
|