Ai2 @allen_ai · 01 Sep 2026
BenchMIRT builds on Item Response Theory (IRT), a technique from psychometrics for measuring abilities from patterns of test responses. The idea: not every question tells you the same amount. Some are harder; some better distinguish stronger models from weaker ones.
623Views
4Likes
0Reposts
1Replies
0Quotes
0Bookmarks
Is that a lot?
0.97×vs this author's median645 views is typical
44Percentile for this authorof 18 recent posts
0.21×vs 10K–100K median2 968 views is typical
0.72%Reachviews ÷ followers
0.80%Engagement rateof viewers reacted