arXivYu Li, Xiuyu Li, Mingyang Yi, Jiaxing Wang, zhangliangxu, Zhaolong Xing, Zhen ChenSun, Jul 5, 2026, 11:59 PM PDT
score 16.7
Selective correction method boosts AI training with off-policy data
Original: Turning Off-Policy Tokens On-Policy: A Plug-in Approach for Improving LLM Alignment
Source: arxiv.org ↗
Writing ELI5 summary…