Significant Improvements over the State of the Art? A Case Study of the MS MARCO Document Ranking Leaderboard

TL;DR

This paper critically examines the evaluation methods used in MS MARCO document ranking leaderboard, highlighting the importance of significance testing and proposing a framework to better assess true improvements over the state of the art.

Contribution

It introduces an evaluation framework that explicitly considers significance testing and avoids metric aggregation, providing more reliable comparisons of IR system performance.

Findings

Current metrics may obscure true performance differences

Significance testing can reveal meaningful improvements

Evaluation methodology impacts perceived progress in IR

Abstract

Leaderboards are a ubiquitous part of modern research in applied machine learning. By design, they sort entries into some linear order, where the top-scoring entry is recognized as the "state of the art" (SOTA). Due to the rapid progress being made in information retrieval today, particularly with neural models, the top entry in a leaderboard is replaced with some regularity. These are touted as improvements in the state of the art. Such pronouncements, however, are almost never qualified with significance testing. In the context of the MS MARCO document ranking leaderboard, we pose a specific question: How do we know if a run is significantly better than the current SOTA? We ask this question against the backdrop of recent IR debates on scale types: in particular, whether commonly used significance tests are even mathematically permissible. Recognizing these potential pitfalls in evaluation methodology, our study proposes an evaluation framework that explicitly treats certain outcomes as distinct and avoids aggregating them into a single-point metric. Empirical analysis of SOTA runs from the MS MARCO document ranking leaderboard reveals insights about how one run can be "significantly better" than another that are obscured by the current official evaluation metric (MRR@100).