RareDiseaseBench Tests AI Agents on Rare Disease Diagnosis

Researchers released RareDiseaseBench, a benchmark of 35 diagnostic workflow tasks for rare diseases affecting some 300 million people, where even the strongest AI model passes only 43.8% of cases.

2026-11-07 ~ 2026-11-07 · 3 related posts

1 near-duplicate retellings: kenbwork