FashionLens — a unified MLLM framework for versatile fashion image retrieval.</p>\n","updatedAt":"2026-05-22T13:53:20.588Z","author":{"_id":"65375039d325b3f02fb4df60","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/65375039d325b3f02fb4df60/gztzXoIci35tqy_E2nAI4.png","fullname":"Haokun Wen","name":"HaokunWen","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"isUserFollowing":false}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.6852503418922424},"editors":["HaokunWen"],"editorAvatarUrls":["https://cdn-avatars.huggingface.co/v1/production/uploads/65375039d325b3f02fb4df60/gztzXoIci35tqy_E2nAI4.png"],"reactions":[],"isReport":false}}],"primaryEmailConfirmed":false,"paper":{"id":"2605.22552","authors":[{"_id":"6a105e8d4cd7a376798ea188","user":{"_id":"65375039d325b3f02fb4df60","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/65375039d325b3f02fb4df60/gztzXoIci35tqy_E2nAI4.png","isPro":false,"fullname":"Haokun Wen","user":"HaokunWen","type":"user","name":"HaokunWen"},"name":"Haokun Wen","status":"claimed_verified","statusLastChangedAt":"2026-05-22T15:58:56.012Z","hidden":false},{"_id":"6a105e8d4cd7a376798ea189","name":"Xuemeng Song","hidden":false},{"_id":"6a105e8d4cd7a376798ea18a","name":"Xinghao Xie","hidden":false},{"_id":"6a105e8d4cd7a376798ea18b","name":"Xiaolin Chen","hidden":false},{"_id":"6a105e8d4cd7a376798ea18c","name":"Xiangyu Zhao","hidden":false},{"_id":"6a105e8d4cd7a376798ea18d","name":"Weili Guan","hidden":false}],"publishedAt":"2026-05-21T00:00:00.000Z","submittedOnDailyAt":"2026-05-22T00:00:00.000Z","title":"FashionLens: Toward Versatile Fashion Image Retrieval via Task-Adaptive Learning","submittedOnDailyBy":{"_id":"65375039d325b3f02fb4df60","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/65375039d325b3f02fb4df60/gztzXoIci35tqy_E2nAI4.png","isPro":false,"fullname":"Haokun Wen","user":"HaokunWen","type":"user","name":"HaokunWen"},"summary":"Fashion image retrieval is a cornerstone of modern e-commerce systems. A unified framework that supports diverse query formats and search intentions is highly desired in practice. However, existing approaches focus on narrow retrieval tasks and do not fully capture such diversity. Therefore, in this work, we aim to develop a unified framework capable of handling diverse realistic fashion retrieval scenarios, achieving truly versatile fashion image retrieval. To establish a data foundation, we first introduce U-FIRE, a comprehensive benchmark that consolidates fragmented fashion datasets into a unified collection, supplemented by two manually curated datasets for testing generalization. Building upon this, we propose FashionLens, a unified framework based on Multimodal Large Language Models. To handle divergent matching objectives, we design a Proposal-Guided Spherical Query Calibrator that dynamically shifts query representations into task-aligned metric spaces via adaptive spherical linear interpolation. Additionally, to mitigate the optimization imbalance caused by varying task complexities and data scales, we develop a Gradient-Guided Adaptive Sampling strategy that automatically re-weights tasks based on realtime learning difficulty and the data scale prior. Experiments on U-FIRE show that FashionLens achieves state-of-the-art performance across diverse retrieval scenarios and generalizes robustly to unseen tasks. The data and code are publicly released at https://github.com/haokunwen/FashionLens.","upvotes":0,"discussionId":"6a105e8d4cd7a376798ea18e","ai_summary":"A unified fashion image retrieval framework is proposed that handles diverse query formats and search intentions through multimodal large language models with adaptive calibration and sampling strategies.","ai_keywords":["Multimodal Large Language Models","Proposal-Guided Spherical Query Calibrator","Gradient-Guided Adaptive Sampling","fashion image retrieval","unified framework","U-FIRE benchmark"]},"canReadDatabase":false,"canManagePapers":false,"canSubmit":false,"hasHfLevelAccess":false,"upvoted":false,"upvoters":[],"acceptLanguages":["en"],"markdownContentUrl":"https://huggingface.co/buckets/huggingchat/papers-content/resolve/2605/2605.22552.md"}">
FashionLens: Toward Versatile Fashion Image Retrieval via Task-Adaptive Learning
Abstract
A unified fashion image retrieval framework is proposed that handles diverse query formats and search intentions through multimodal large language models with adaptive calibration and sampling strategies.
AI-generated summary
Fashion image retrieval is a cornerstone of modern e-commerce systems. A unified framework that supports diverse query formats and search intentions is highly desired in practice. However, existing approaches focus on narrow retrieval tasks and do not fully capture such diversity. Therefore, in this work, we aim to develop a unified framework capable of handling diverse realistic fashion retrieval scenarios, achieving truly versatile fashion image retrieval. To establish a data foundation, we first introduce U-FIRE, a comprehensive benchmark that consolidates fragmented fashion datasets into a unified collection, supplemented by two manually curated datasets for testing generalization. Building upon this, we propose FashionLens, a unified framework based on Multimodal Large Language Models. To handle divergent matching objectives, we design a Proposal-Guided Spherical Query Calibrator that dynamically shifts query representations into task-aligned metric spaces via adaptive spherical linear interpolation. Additionally, to mitigate the optimization imbalance caused by varying task complexities and data scales, we develop a Gradient-Guided Adaptive Sampling strategy that automatically re-weights tasks based on realtime learning difficulty and the data scale prior. Experiments on U-FIRE show that FashionLens achieves state-of-the-art performance across diverse retrieval scenarios and generalizes robustly to unseen tasks. The data and code are publicly released at https://github.com/haokunwen/FashionLens.
Community
FashionLens — a unified MLLM framework for versatile fashion image retrieval.
Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.
Tap or paste here to upload images
Cite arxiv.org/abs/2605.22552 in a model README.md to link it from this page.
Cite arxiv.org/abs/2605.22552 in a dataset README.md to link it from this page.
Cite arxiv.org/abs/2605.22552 in a Space README.md to link it from this page.
Discussion (0)
Sign in to join the discussion. Free account, 30 seconds — email code or GitHub.
Sign in →No comments yet. Sign in and be the first to say something.