2601.21343v1
arXiv License
Meta FAIR
自我改進式預訓練：利用後訓練模型來預訓練更好的模型
Self-Improving Pretraining using post-trained models to pretrain better models
Ellen Xiaoqing Tan, Jack Lanchantin, Shehzaad Dhuliawala, Danwei Li, Thao Nguyen, Jing Xu, Ping Yu, ...
預訓練強化學習安全性事實性推理中期訓練LLM
Liang-Hsun Huang 許願