Transparent Human Evaluation for Image Captioning

Anonymous

Paper Link: https://openreview.net/forum?id=c6SmGqmGCjY

Paper Type: Long paper (up to eight pages of content + unlimited references and appendices)

Abstract: We establish THumB, a rubric-based human evaluation protocol for image captioning models. Our scoring rubrics and their definitions are carefully developed based on machine- and human-generated captions on the MSCOCO dataset. Each caption is evaluated along two main dimensions in a tradeoff (precision and recall) as well as other aspects that measure the text quality (fluency, conciseness, and inclusive language). Our evaluations demonstrate several critical problems of the current evaluation practice. Human-generated captions show substantially higher quality than machine-generated ones, especially in coverage of salient information (i.e., recall), while most automatic metrics say the opposite. Our rubric-based results reveal that CLIPScore, a recent metric that uses image features, better correlates with human judgments than conventional text-only metrics because it is more sensitive to recall. We hope that this work will promote a more transparent evaluation protocol for image captioning and its automatic metrics.

Presentation Mode: This paper will be presented in person in Seattle

Copyright Consent Signature (type Name Or NA If Not Transferrable): Jungo Kasai

Copyright Consent Name And Address: Paul G. Allen School of Computer Science & Engineering. 185 E Stevens Way NE, Seattle, WA 98195.