generalizability-theory

Tag

Cards List
#generalizability-theory

@dair_ai: There is much less signal in agent leaderboards than the rankings imply. A four-facet Generalizability Theory decomposi…

X AI KOLs Timeline · 2026-08-13 Cached

Researchers show agent leaderboards rank task specialization rather than capability, with agent main effects accounting for under 3% of variance across three benchmarks, and propose a Deployment Decision Reliability framework.

0 favorites 0 likes
#generalizability-theory

Deployment Decision Reliability: A Generalizability-Theory Framework for Sizing Long-Horizon Agent Evaluations

arXiv cs.AI · 2026-08-13 Cached

This paper applies Generalizability Theory to agent benchmarks, showing leaderboards rank specialization rather than capability, and proposes a framework (DDR) for sizing reliable deployment evaluations.

0 favorites 0 likes
← Back to home

Submit Feedback