# ZooClaw-FashionSigLIP2: Distilled Fine-tuning for Robust Fashion Retrieval

Source: [ZooClaw-FashionSigLIP2: Distilled Fine-tuning for Robust Fashion Retrieval](<https://www.sota2.com/research/paper/6a41fc23206542251d7ccbe9>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-07-03T06:00:01.668Z

Authors: Siqiao Xue, Chunxue Xu

Published: 2026-06-26T04:13:32.000Z

## Abstract

Adapting a foundation vision-language encoder to a specialized retrieval task creates a fundamental tradeoff: gains on the target distribution come at the cost of the foundation model's broad generalization, and fashion retrieval is a stringent instance of this problem. We present ZooClaw-FashionSigLIP2, a fashion-specialized SigLIP2-base model that resolves this tradeoff with a simple recipe -- full fine-tuning with knowledge distillation on curated in-domain data, followed by \\wiseft~\\citep{wortsman2022wiseft} weight interpolation with the base model -- and outperforms LoRA, larger backbones (up to 1B parameters), and external training data. Under fair evaluation, ZooClaw-FashionSigLIP2 outperforms all baselines on every benchmark in our suite. In addition, we release ZooClaw-Fashion, a new high-quality fashion retrieval benchmark, and a systematic quality analysis of widely-used benchmarks that exposes and mitigates structural biases in their public ground truth. We open-source the model weights and all evaluation artifacts to facilitate future research.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Text-to-image fashion retrieval on ZooClaw-Fashion long query (test)](<https://www.sota2.com/research/sota/text-to-image-fashion-retrieval-on-zooclaw-fashion-long-query-test.md>): Text-to-image fashion retrieval; ZooClaw-Fashion long query (test); R@1: 44.9.

- [Text-to-Image Retrieval on ZooClaw-Fashion short query (test)](<https://www.sota2.com/research/sota/text-to-image-retrieval-on-zooclaw-fashion-short-query-test.md>): Text-to-image fashion retrieval; ZooClaw-Fashion short query (test); R@1: 42.3.

- [Text-to-Image Retrieval on Fashion200k (R@10, MRR)](<https://www.sota2.com/research/sota/text-to-image-retrieval-on-fashion200k-r-at-10-mrr.md>): Text-to-image fashion retrieval; Fashion200k; R@10: 28.6.

- [Text-to-Image Fashion Retrieval on H&M](<https://www.sota2.com/research/sota/text-to-image-fashion-retrieval-on-h-and-m.md>): Text-to-image fashion retrieval; H&M; R@10: 13.6.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2606.27708>)

[PDF](<https://arxiv.org/pdf/2606.27708.pdf>)