Polos: Multimodal Metric Learning from Human Feedback for Image Captioning

Yuiga Wada, Kanta Kaneda, Daichi Saito, Komei Sugiura; Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2024, pp. 13559-13568

Establishing an automatic evaluation metric that closely aligns with human judgments is essential for effectively developing image captioning models. Recent data-driven metrics have demonstrated a stronger correlation with human judgments than classic metrics such as CIDEr; however they lack sufficient capabilities to handle hallucinations and generalize across diverse images and texts partially because they compute scalar similarities merely using embeddings learned from tasks unrelated to image captioning evaluation. In this study we propose Polos a supervised automatic evaluation metric for image captioning models. Polos computes scores from multimodal inputs using a parallel feature extraction mechanism that leverages embeddings trained through large-scale contrastive learning. To train Polos we introduce Multimodal Metric Learning from Human Feedback (M2LHF) a framework for developing metrics based on human feedback. We constructed the Polaris dataset which comprises 131K human judgments from 550 evaluators which is approximately ten times larger than standard datasets. Our approach achieved state-of-the-art performance on Composite Flickr8K-Expert Flickr8K-CF PASCAL-50S FOIL and the Polaris dataset thereby demonstrating its effectiveness and robustness.