{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/reason-before-retrieve-one-stage-reflective","title":"Reason-before-Retrieve: One-Stage Reflective Chain-of-Thoughts for Training-Free Zero-Shot Composed Image Retrieval","arxiv_id":"2412.11077","date":"2024-12-15","proceeding":"CVPR 2025 1","authors":["Yuanmin Tang","Xiaoting Qin","Jue Zhang","Jing Yu","Gaopeng Gou","Gang Xiong","Qingwei Ling","Saravan Rajmohan","Dongmei Zhang","Qi Wu"],"abstract":"Composed Image Retrieval (CIR) aims to retrieve target images that closely resemble a reference image while integrating user-specified textual modifications, thereby capturing user intent more precisely. Existing training-free zero-shot CIR (ZS-CIR) methods often employ a two-stage process: they first generate a caption for the reference image and then use Large Language Models for reasoning to obtain a target description. However, these methods suffer from missing critical visual details and limited reasoning capabilities, leading to suboptimal retrieval performance. To address these challenges, we propose a novel, training-free one-stage method, One-Stage Reflective Chain-of-Thought Reasoning for ZS-CIR (OSrCIR), which employs Multimodal Large Language Models to retain essential visual information in a single-stage reasoning process, eliminating the information loss seen in two-stage methods. Our Reflective Chain-of-Thought framework further improves interpretative accuracy by aligning manipulation intent with contextual cues from reference images. OSrCIR achieves performance gains of 1.80% to 6.44% over existing training-free methods across multiple tasks, setting new state-of-the-art results in ZS-CIR and enhancing its utility in vision-language applications. Our code will be available at https://github.com/Pter61/osrcir2024/.","url_abs":"https://arxiv.org/abs/2412.11077v3","url_pdf":"https://arxiv.org/pdf/2412.11077v3.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"reason-before-retrieve-one-stage-reflective","repo_url":"https://github.com/Pter61/osrcir","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"image-retrieval","task_name":"Image Retrieval"},{"task_slug":"retrieval","task_name":"Retrieval"},{"task_slug":"zero-shot-composed-image-retrieval-zs-cir","task_name":"Zero-Shot Composed Image Retrieval (ZS-CIR)"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/zero-shot-composed-image-retrieval-zs-cir-on","task":"Zero-Shot Composed Image Retrieval (ZS-CIR)","dataset":"CIRCO","model":"OSrCIR (CLIP G/14)","rank_in_archive_order":11,"of":43,"metrics":{"mAP@10":"31.14"},"uses_additional_data":false},{"leaderboard":"/sota/zero-shot-composed-image-retrieval-zs-cir-on","task":"Zero-Shot Composed Image Retrieval (ZS-CIR)","dataset":"CIRCO","model":"OSrCIR (CLIP L/14)","rank_in_archive_order":18,"of":43,"metrics":{"mAP@10":"25.33"},"uses_additional_data":false},{"leaderboard":"/sota/zero-shot-composed-image-retrieval-zs-cir-on","task":"Zero-Shot Composed Image Retrieval (ZS-CIR)","dataset":"CIRCO","model":"OSrCIR (CLIP B/32)","rank_in_archive_order":25,"of":43,"metrics":{"mAP@10":"19.17"},"uses_additional_data":false},{"leaderboard":"/sota/zero-shot-composed-image-retrieval-zs-cir-on-1","task":"Zero-Shot Composed Image Retrieval (ZS-CIR)","dataset":"CIRR","model":"OSrCIR (CLIP G/14)","rank_in_archive_order":7,"of":47,"metrics":{"R@1":"37.26","R@10":"77.33","R@5":"67.25"},"uses_additional_data":false},{"leaderboard":"/sota/zero-shot-composed-image-retrieval-zs-cir-on-1","task":"Zero-Shot Composed Image Retrieval (ZS-CIR)","dataset":"CIRR","model":"OSrCIR (CLIP L/14)","rank_in_archive_order":17,"of":47,"metrics":{"R@1":"29.45","R@10":"69.86","R@5":"57.68"},"uses_additional_data":false},{"leaderboard":"/sota/zero-shot-composed-image-retrieval-zs-cir-on-1","task":"Zero-Shot Composed Image Retrieval (ZS-CIR)","dataset":"CIRR","model":"OSrCIR (CLIP B/32)","rank_in_archive_order":20,"of":47,"metrics":{"R@1":"25.42","R@10":"68.19","R@5":"54.54"},"uses_additional_data":false},{"leaderboard":"/sota/zero-shot-composed-image-retrieval-zs-cir-on-2","task":"Zero-Shot Composed Image Retrieval (ZS-CIR)","dataset":"Fashion IQ","model":"OSrCIR (CLIP G/14)","rank_in_archive_order":8,"of":41,"metrics":{"(Recall@10+Recall@50)/2":"47.34"},"uses_additional_data":false},{"leaderboard":"/sota/zero-shot-composed-image-retrieval-zs-cir-on-2","task":"Zero-Shot Composed Image Retrieval (ZS-CIR)","dataset":"Fashion IQ","model":"OSrCIR (CLIP B/32)","rank_in_archive_order":18,"of":41,"metrics":{"(Recall@10+Recall@50)/2":"42.87"},"uses_additional_data":false},{"leaderboard":"/sota/zero-shot-composed-image-retrieval-zs-cir-on-2","task":"Zero-Shot Composed Image Retrieval (ZS-CIR)","dataset":"Fashion IQ","model":"OSrCIR (CLIP L/14)","rank_in_archive_order":19,"of":41,"metrics":{"(Recall@10+Recall@50)/2":"42.82"},"uses_additional_data":false},{"leaderboard":"/sota/zero-shot-composed-image-retrieval-zs-cir-on-11","task":"Zero-Shot Composed Image Retrieval (ZS-CIR)","dataset":"GeneCIS","model":"OSrCIR (CLIP G/14)","rank_in_archive_order":1,"of":11,"metrics":{" A-R@1":"19.6"},"uses_additional_data":false},{"leaderboard":"/sota/zero-shot-composed-image-retrieval-zs-cir-on-11","task":"Zero-Shot Composed Image Retrieval (ZS-CIR)","dataset":"GeneCIS","model":"OSrCIR (CLIP L/14)","rank_in_archive_order":2,"of":11,"metrics":{" A-R@1":"17.9"},"uses_additional_data":false},{"leaderboard":"/sota/zero-shot-composed-image-retrieval-zs-cir-on-11","task":"Zero-Shot Composed Image Retrieval (ZS-CIR)","dataset":"GeneCIS","model":"OSrCIR (CLIP B/32)","rank_in_archive_order":3,"of":11,"metrics":{" A-R@1":"17.4"},"uses_additional_data":false}],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}