$T^5Score$: A Methodology for Automatically Assessing the Quality of LLM Generated Multi-Document Topic Sets

TL;DR

This paper introduces $T^5Score$, a new evaluation methodology for assessing the quality of multi-document topic sets generated by LLMs, addressing the limitations of existing practices and enabling reliable, high-agreement assessments.

Contribution

The paper presents $T^5Score$, a novel, decompositional evaluation framework for LLM-generated topics that improves reliability and inter-annotator agreement.

Findings

$T^5Score$ achieves high inter-annotator agreement.

It effectively decomposes topic quality into measurable aspects.

Experimental results validate its applicability across datasets.

Abstract

Using LLMs for Multi-Document Topic Extraction has recently gained popularity due to their apparent high-quality outputs, expressiveness, and ease of use. However, most existing evaluation practices are not designed for LLM-generated topics and result in low inter-annotator agreement scores, hindering the reliable use of LLMs for the task. To address this, we introduce $T^5Score$, an evaluation methodology that decomposes the quality of a topic set into quantifiable aspects, measurable through easy-to-perform annotation tasks. This framing enables a convenient, manual or automatic, evaluation procedure resulting in a strong inter-annotator agreement score. To substantiate our methodology and claims, we perform extensive experimentation on multiple datasets and report the results.