TextClass Benchmark: A Continuous Elo Rating of LLMs in Social Sciences

TL;DR

The TextClass Benchmark introduces a continuous, dynamic evaluation system for LLMs in social science text classification, using Elo ratings to track performance across multiple languages and domains.

Contribution

It presents a novel ongoing benchmarking framework with Elo-based ratings, accommodating multiple languages, domains, and test set updates for fair comparison of LLMs.

Findings

First cycle results on incivility data across four languages.

Inclusion of multiple languages and social science topics.

Demonstration of the Meta-Elo system for aggregated rankings.

Abstract

The TextClass Benchmark project is an ongoing, continuous benchmarking process that aims to provide a comprehensive, fair, and dynamic evaluation of LLMs and transformers for text classification tasks. This evaluation spans various domains and languages in social sciences disciplines engaged in NLP and text-as-data approach. The leaderboards present performance metrics and relative ranking using a tailored Elo rating system. With each leaderboard cycle, novel models are added, fixed test sets can be replaced for unseen, equivalent data to test generalisation power, ratings are updated, and a Meta-Elo leaderboard combines and weights domain-specific leaderboards. This article presents the rationale and motivation behind the project, explains the Elo rating system in detail, and estimates Meta-Elo across different classification tasks in social science disciplines. We also present a snapshot of the first cycle of classification tasks on incivility data in Chinese, English, German and Russian. This ongoing benchmarking process includes not only additional languages such as Arabic, Hindi, and Spanish but also a classification of policy agenda topics, misinformation, among others.