Bidimensional Leaderboards: Generate and Evaluate Language Hand in Hand

TL;DR

This paper introduces bidimensional leaderboards (Billboards) that simultaneously evaluate language generation models and metrics, enabling the development of better evaluation tools that correlate more closely with human judgments.

Contribution

The paper proposes a novel bidimensional leaderboard framework that tracks progress in both language models and evaluation metrics, fostering mutual improvement and more accurate assessment.

Findings

Ensemble metrics can outperform individual metrics in evaluation.

Most automatic metrics tend to overrate machine-generated content compared to human.

Updating metrics is crucial as models become more human-like.

Abstract

Natural language processing researchers have identified limitations of evaluation methodology for generation tasks, with new questions raised about the validity of automatic metrics and of crowdworker judgments. Meanwhile, efforts to improve generation models tend to depend on simple n-gram overlap metrics (e.g., BLEU, ROUGE). We argue that new advances on models and metrics should each more directly benefit and inform the other. We therefore propose a generalization of leaderboards, bidimensional leaderboards (Billboards), that simultaneously tracks progress in language generation models and metrics for their evaluation. Unlike conventional unidimensional leaderboards that sort submitted systems by predetermined metrics, a Billboard accepts both generators and evaluation metrics as competing entries. A Billboard automatically creates an ensemble metric that selects and linearly combines a few metrics based on a global analysis across generators. Further, metrics are ranked based on their correlation with human judgments. We release four Billboards for machine translation, summarization, and image captioning. We demonstrate that a linear ensemble of a few diverse metrics sometimes substantially outperforms existing metrics in isolation. Our mixed-effects model analysis shows that most automatic metrics, especially the reference-based ones, overrate machine over human generation, demonstrating the importance of updating metrics as generation models become stronger (and perhaps more similar to humans) in the future. Our project website is available at https://nlp.cs.washington.edu/billboard/.