Concept detail
human evaluation at scale
0 tagged videos · 1 connection videos · degree 0
Publication timeline
- 2026-081 video
Source videos
Observed neighbors
reward hacking and bias detection
trusted tester feedback catches systematic artifacts and nuanced aesthetic blind spots that automated reward models miss.
1 source videos
SOTA Generative Media Panel — Dumitru Erhan, Shane Gu & Nicole Brichtova, Google DeepMindRelated tagged insights
No core insights are tagged with this concept.