Roughly 12% of ImageNet validation labels are just wrong, researcher says
RexDouglass · x · 2026-09-03
In a related post, klaaracz notes ImageNet was built when the question was "cat or dog", but we now ask "husky or malamute" — and the labels never caught up. Roughly 12% of the original validation labels are simply wrong.
More from Research
- AI reads chest CTs to score each organ's age, validated on 35,000 scans — rand_longevity · 2026-09-03
- Eigenfaces revisited: interactive demos of the pioneering computer vision technique — CSProfKGD · 2026-09-03
- Eigenfaces nostalgia: CV professor revisits the technique that defined early face recognition — CSProfKGD · 2026-09-03
- HCOMP 2026 + CI 2026 joint conference registration open; student early-bird at $300 — windx0303 · 2026-09-03
- Three-Paper Series Decomposes Human-Like RSI into ASPIRE, S³Gym and HarnessDev — teortaxesTex · 2026-09-03
- Noematrix's Noe-0 trains embodied models with zero teleoperation data — jiqizhixin · 2026-09-03