TY  - RPRT
TI  - Bridging the Training-Inference Gap in LLMs by Leveraging Self-Generated Tokens
AU  - Zhepeng Cen
AU  - Yao Liu
AU  - Siliang Zeng
AU  - Pratik Chaudhari
AU  - Huzefa Rangwala
AU  - George Karypis
AU  - Rasool Fakoor
PY  - 2025
UR  - https://arxiv.org/abs/2410.14655
ID  - 2410.14655
ER  -