Saved in:
Bibliographic Details
Main Authors: Ramaswamy, Ashwin, Demeure, Nestor, Rrapaj, Ermal
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2509.23510
Tags: Add Tag
No Tags, Be the first to tag this record!
Table of Contents:
  • New large language models (LLMs) are being released every day. Some perform significantly better or worse than expected given their parameter count. Therefore, there is a need for a method to independently evaluate models. The current best way to evaluate a model is to measure its Elo score by comparing it to other models in a series of contests - an expensive operation since humans are ideally required to compare LLM outputs. We observe that when an LLM is asked to judge such contests, the consistency with which it selects a model as the best in a matchup produces a metric that is 91% correlated with its own human-produced Elo score. This provides a simple proxy for Elo scores that can be computed cheaply, without any human data or prior knowledge.