Dynamically Allocating Evaluation Effort for Model Ranking

arXiv:2608.03437 · cs.CL, cs.LG · Submitted 2026-08-04 · Read on arXiv

cs.CL, cs.LG

Submitted: 2026-08-04

Updated: 2026-08-29

Code: https://github.com/zouharvi/evaluation-bandit

License: http://creativecommons.org/licenses/by/4.0/

The gist: While human evaluation is the gold standard in many NLP tasks, it suffers from prohibitive costs and poor scalability.

Terminology

Abstract

While human evaluation is the gold standard in many NLP tasks, it suffers from prohibitive costs and poor scalability. When identifying top-performing models, typical evaluation protocols waste effort by exhaustively evaluating all models on the entire benchmark, a safe but inefficient approach. In this work, we formalize multi-model human evaluation as a best-arm identification problem in a multi-armed bandit setup with correlated arms, where pulling an arm corresponds to human-evaluating a model. By sampling adaptively based on the intermediate model rankings obtained on the samples so far, we can focus the annotation budget on the most competitive models. We prove the optimality of the proposed algorithms and show that it improves discrimination between top-performing models. This makes evaluations faster, cheaper and more aligned with large-scale competition evaluation goals.

Sources

Related papers