NoxionAI/comprehension-score

2 best points · 0 comments · 1 Hacker News post · View on GitHub

Hacker News discussions

• LLM evals test outputs. Rarely whether the model understood first (2 points, 0 comments, by noxion, 2026-03-20)