| 1 |
Surveying the Effects of Quality, Diversity, and Complexity in Synthetic Data From Large Language Models |
通过评估合成数据的质量、多样性和复杂性，深入分析大语言模型生成合成数据的影响。 |
reinforcement learning large language model |
|
|
| 2 |
Tight PAC-Bayesian Risk Certificates for Contrastive Learning |
提出基于PAC-Bayes的对比学习风险证书，解决SimCLR框架下的泛化性保证问题。 |
representation learning contrastive learning foundation model |
|
|
| 3 |
PathletRL++: Optimizing Trajectory Pathlet Extraction and Dictionary Formation via Reinforcement Learning |
PathletRL++：通过强化学习优化轨迹Pathlet提取和字典构建，提升轨迹数据表示效率。 |
reinforcement learning deep reinforcement learning |
|
|
| 4 |
Cluster Specific Representation Learning |
提出聚类特定表示学习框架，提升下游任务的泛化性能 |
representation learning contrastive learning |
|
|
| 5 |
Inverse Delayed Reinforcement Learning |
提出逆延迟强化学习框架，从受延迟扰动的专家轨迹中提取奖励特征并恢复策略。 |
reinforcement learning inverse reinforcement learning |
|
|
| 6 |
Hyper: Hyperparameter Robust Efficient Exploration in Reinforcement Learning |
提出Hyper算法，解决强化学习中探索策略对超参数敏感的问题 |
reinforcement learning |
|
|
| 7 |
AI-Driven Day-to-Day Route Choice |
提出基于LLM的出行者建模框架LLMTraveler，用于模拟日常路径选择行为 |
reinforcement learning large language model |
|
|