Qistas: A Pre-Registered, Portable Protocol for Fairness Evaluation of Large Language Models

A proposed approach aims to make LLM fairness evaluations portable and preregistered.

Fairness benchmarks for large language models (LLMs) are overwhelmingly English-born and American in their social categories; translated versions carry those categories into societies where they do not apply, and almost none fix their hypotheses before the models are run.

ArXiv CS.CL2026-11-02