English

An Efficient Post-hoc Framework for Reducing Task Discrepancy of Text Encoders for Composed Image Retrieval

Computer Vision and Pattern Recognition 2025-03-19 v2 Information Retrieval

Abstract

Composed Image Retrieval (CIR) aims to retrieve a target image based on a reference image and conditioning text, enabling controllable image searches. The mainstream Zero-Shot (ZS) CIR methods bypass the need for expensive training CIR triplets by projecting image embeddings into the text token embedding space, forming a composed query for retrieval. However, we highlight an inherent limitation in these projection-based CIR: a task discrepancy of text encoders between the original pre-training task of the encoders (text \leftrightarrow image) and the target CIR task (image + text \leftrightarrow image), which potentially negatively impacts CIR performance. To reduce such a discrepancy, a naive solution would be to train both image and text encoders with CIR triplets in a supervised manner. Instead, we introduce Reducing Task Discrepancy of Text Encoders (RTD), an efficient text-only post-hoc framework that complements projection-based CIR methods. We devise a novel target-anchored text contrastive learning designed to enhance the capability of the text encoder for CIR. We also propose two key enhancements: (1) a hard negative-based refined batch sampling strategy and (2) a refined concatenation scheme to further mitigate training-inference discrepancy. Integrating RTD into state-of-the-art projection-based methods achieves performance comparable to, or even surpassing, resource-intensive state-of-the-art synthetic CIR triplet-based approaches only with 23 minutes of additional training on 4 A100 GPUs (up to 100×100\times faster in training). Our code will be available upon acceptance.

Keywords

Cite

@article{arxiv.2406.09188,
  title  = {An Efficient Post-hoc Framework for Reducing Task Discrepancy of Text Encoders for Composed Image Retrieval},
  author = {Jaeseok Byun and Seokhyeon Jeong and Wonjae Kim and Sanghyuk Chun and Taesup Moon},
  journal= {arXiv preprint arXiv:2406.09188},
  year   = {2025}
}

Comments

22 pages

R2 v1 2026-06-28T17:04:40.637Z