原书出处
Designing Large Language Model Applications
we do not have access to ground truth. But for summarization applications, we do have access to the ground truth. Therefore, we can verify factuality by taking each sentence in the summary and checking whether, given the source text, one can logically conclude the statement in the summary. This can be framed as a natural language inference (NLI) problem, which is a standard NLP task.
原书定位详情
来源:a19c7933-4eda-4176-aebd-30fe7e1ad292
正文位置:extracted/Designing Large Language Model Applications/_epub-src/OEBPS/ch10.html
EPUB SHA-256:1b4d2347fec2f949fcddfbaf4fad597a1ae20887a096cfae64c59a619b4781bf