NEWVectors or files. Pick a path.Start →
    Back to Videos

    Composed Image Retrieval With No Vision Model: 44% R@1 Zero-Shot

    63
    Retrieval & Search Research
    Mixpeek Team
    July 20, 2026

    Summary

    Composed image retrieval handles queries like 'this jacket, but in red, with a hood.' A University of Tokyo team does it with images represented purely as text: a caption plus an explicit attribute checklist at ingest, an LLM to merge the modification, plain text retrieval, attribute scoring, and an LLM reranker to verify the change happened. Zero-shot: 44.04% R@1 on CIRR, +8.79 over prior methods. Paper: arxiv.org/abs/2607.12621

    composed-image-retrievalvisual-searchcirllm-rerankingmultimodal-search

    About this video

    Composed image retrieval handles queries like 'this jacket, but in red, with a hood.' A University of Tokyo team does it with images represented purely as text: a caption plus an explicit attribute checklist at ingest, an LLM to merge the modification, plain text retrieval, attribute scoring, and an LLM reranker to verify the change happened. Zero-shot: 44.04% R@1 on CIRR, +8.79 over prior methods. Paper: arxiv.org/abs/2607.12621

    Frequently Asked Questions