In the paper, they say they used the most recent 56 cases (from 2024–2025) as a ... | Hacker News

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

		brandonb 5 months ago \| parent \| context \| favorite \| on: The Path to Medical Superintelligence In the paper, they say they used the most recent 56 cases (from 2024–2025) as a holdout set. The majority of those cases happened after the o4 training cutoff of May 31, 2024.

miraculixx 5 months ago [–]

Are these 56 cases distinct from all other cases in the data?

FlyingLawnmower 5 months ago | [–]

Yes. They are about entirely different patient reports.

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