CLAIR: Evaluating Image Captions with Large Language Models

David Chan, Suzanne Petryk, Joseph Gonzalez, Trevor Darrell, John Canny


Abstract
The evaluation of machine-generated image captions poses an interesting yet persistent challenge. Effective evaluation measures must consider numerous dimensions of similarity, including semantic relevance, visual structure, object interactions, caption diversity, and specificity. Existing highly-engineered measures attempt to capture specific aspects, but fall short in providing a holistic score that aligns closely with human judgments. Here, we propose CLAIR, a novel method that leverages the zero-shot language modeling capabilities of large language models (LLMs) to evaluate candidate captions. In our evaluations, CLAIR demonstrates a stronger correlation with human judgments of caption quality compared to existing measures. Notably, on Flickr8K-Expert, CLAIR achieves relative correlation improvements over SPICE of 39.6% and over image-augmented methods such as RefCLIP-S of 18.3%. Moreover, CLAIR provides noisily interpretable results by allowing the language model to identify the underlying reasoning behind its assigned score.
Anthology ID:
2023.emnlp-main.841
Original:
2023.emnlp-main.841v1
Version 2:
2023.emnlp-main.841v2
Volume:
Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing
Month:
December
Year:
2023
Address:
Singapore
Editors:
Houda Bouamor, Juan Pino, Kalika Bali
Venue:
EMNLP
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
13638–13646
Language:
URL:
https://aclanthology.org/2023.emnlp-main.841
DOI:
10.18653/v1/2023.emnlp-main.841
Bibkey:
Cite (ACL):
David Chan, Suzanne Petryk, Joseph Gonzalez, Trevor Darrell, and John Canny. 2023. CLAIR: Evaluating Image Captions with Large Language Models. In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing, pages 13638–13646, Singapore. Association for Computational Linguistics.
Cite (Informal):
CLAIR: Evaluating Image Captions with Large Language Models (Chan et al., EMNLP 2023)
Copy Citation:
PDF:
https://aclanthology.org/2023.emnlp-main.841.pdf
Video:
 https://aclanthology.org/2023.emnlp-main.841.mp4