Anecdata seems quite valid for LLM comparison when trying to evaluate 'usefullne... | Hacker News

Hacker Newsnew | past | comments | ask | show | jobs | submit

		Chilko on July 25, 2024 \| parent \| context \| favorite \| on: Large Enough Anecdata seems quite valid for LLM comparison when trying to evaluate 'usefullness' for users. The lmsys chat leaderboard is literally just mass anecdata.

lostmsu on July 25, 2024 [–]

Yes, "mass anecdata" + blind collection is usually called "data".

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact