New Metrics to Encourage Innovation and Diversity in Information Retrieval Approaches

TL;DR

This paper introduces new IR evaluation metrics designed to promote diversity and innovation by rewarding systems that explore divergent strategies and find unique relevant documents, thus encouraging more radical approaches.

Contribution

It proposes a novel class of IR metrics that incentivize exploration of diverse, high-risk strategies, contrasting with traditional metrics focused on user experience.

Findings

Metrics change system rankings to favor innovative approaches.

Metrics reward systems that discover rare, relevant documents.

Metrics demonstrate higher stability and discriminative power.

Abstract

In evaluation campaigns, participants often explore variations of popular, state-of-the-art baselines as a low-risk strategy to achieve competitive results. While effective, this can lead to local "hill climbing" rather than more radical and innovative departure from standard methods. Moreover, if many participants build on similar baselines, the overall diversity of approaches considered may be limited. In this work, we propose a new class of IR evaluation metrics intended to promote greater diversity of approaches in evaluation campaigns. Whereas traditional IR metrics focus on user experience, our two "innovation" metrics instead reward exploration of more divergent, higher-risk strategies finding relevant documents missed by other systems. Experiments on four TREC collections show that our metrics do change system rankings by rewarding systems that find such rare, relevant documents. This result is further supported by a controlled, synthetic data experiment, and a qualitative analysis. In addition, we show that our metrics achieve higher evaluation stability and discriminative power than the standard metrics we modify. To support reproducibility, we share our source code.