Metraipolis › archive › dc_bb1f9cae6780
Preserve-and-Compose Training for Composed Image Retrieval — arXiv:2609.31202
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.31202
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-08T03:12:44.442Z
- SHA-256 of the archived text
aaf19d6635fe0a7252ae8a0ffdb48e54299a086c0151288e431cd96c59bea8a5- Stable URL of this page
https://metraipolis.com/documento/dc_bb1f9cae6780
How to cite
Preserve-and-Compose Training for Composed Image Retrieval — arXiv:2609.31202 — arXiv (en), https://arxiv.org/abs/2609.31202 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_bb1f9cae6780 (Metraipolis archive dc_bb1f9cae6780, sha256 aaf19d6635fe0a7252ae8a0ffdb48e54299a086c0151288e431cd96c59bea8a5)
BibTeX
@misc{dcbb1f9cae6780,
title = {Preserve-and-Compose Training for Composed Image Retrieval — arXiv:2609.31202},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.31202},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_bb1f9cae6780; sha256 aaf19d6635fe0a7252ae8a0ffdb48e54299a086c0151288e431cd96c59bea8a5},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.31202. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Composed image retrieval (CIR) aims to retrieve images that satisfy a user-specified modification while preserving relevant visual content from a reference image. Collecting target images for this purpose is costly, motivating zero-shot CIR methods that instead use target captions as supervision. However, target captions may omit source details that should be preserved. We therefore propose, Preserve-and-Compose Training, which complements target-caption supervision with visual evidence from the source image. PACT learns from image--text--text (ITT) triplets without target images or gallery updates, aligning composed queries with target captions while preserving source evidence through visual supervision. We further introduce Chord scoring, which combines target similarity with source-relative directional agreement in the frozen image space. Results across four ZS-CIR benchmarks show that combining target-caption supervision with source-image evidence leads to strong retrieval performance across datasets, backbone scales, and external galleries. The code is available on this https URL . Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-09-28 in cs.CV). Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV). Pagina del lavoro: https://arxiv.org/abs/2609.31202 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-08T02:50:19.090Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-09-28?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).