[ICASSP'26] HINT: Composed Image Retrieval with Dual-Path Compositional Contextualized Network

Mingyu Zhang1, Zixu Li1, Zhiwei Chen1, Zhiheng Fu1, Xiaowei Zhu1, Jiajia Nie1, Yinwei Wei1, Yupeng Hu1*,
1School of Software, Shandong University
*Corresponding author.
Paper Code

Abstract

MY ALT TEXT

Hint Discriminative Advantages

MY ALT TEXT

An example of the CIR task and HINT's main idea.


HINT: dual-patH composItional coNtextualized neTwork

MY ALT TEXT

HINT framework: (a) Dual Context Extraction, (b) Quantification of Contextual Relevance, (c) Dual-Path Consistency Constraints.

Experiment

MY ALT TEXT

Performance comparison on FashionIQ and CIRR relative to R@k(%). The overall best results are in bold, while the second-best results are underlined. The Avg metric in CIRR denotes (R@5 + Rsubset@1) / 2.

MY ALT TEXT

Ablation study on FashionIQ and CIRR datasets. We compute Avg-R@10, Avg-R@50 for FashionIQ, and Avg (mean of R@5 and Rsubset@1) for CIRR, respectively

MY ALT TEXT

Case study on CIRR and FashionIQ.

BibTeX


        @article{zhang2026hint,
          title={Hint: Composed image retrieval with dual-path compositional contextualized network},
          author={Zhang, Mingyu and Li, Zixu and Chen, Zhiwei and Fu, Zhiheng and Zhu, Xiaowei and Nie, Jiajia and Wei, Yinwei and
          Hu, Yupeng},
          journal={arXiv preprint arXiv:2603.26341},
          year={2026}
        }