1 paper · 1 filter
Chen Shen, Wei Cheng, Jingyue Yang +3
The proficiency of Large Language Models (LLMs) in coding tasks is often a reflection of their extensive pre-training corpora, which typically collapses when confronted with previo…