STABLEVAL models latent correctness and annotator confusion to deliver more stable and uncertainty-aware AI system rankings than majority-vote aggregation.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
STABLEVAL: Disagreement-Aware and Stable Evaluation of AI Systems
STABLEVAL models latent correctness and annotator confusion to deliver more stable and uncertainty-aware AI system rankings than majority-vote aggregation.