热点
"人类偏好对齐" 相关文章
Difficulty-Based Preference Data Selection by DPO Implicit Reward Gap
cs.AI updates on arXiv.org 2025-08-07T04:12:51.000000Z
Gradient-Adaptive Policy Optimization: Towards Multi-Objective Alignment of Large Language Models
cs.AI updates on arXiv.org 2025-07-03T04:07:28.000000Z