On the Provable Importance of Gradients for Language-Assisted Image Clustering

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Peng, Bo, Lu, Jie, Zhang, Guangquan, Fang, Zhen
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914589125378048
author Peng, Bo
Lu, Jie
Zhang, Guangquan
Fang, Zhen
author_facet Peng, Bo
Lu, Jie
Zhang, Guangquan
Fang, Zhen
contents This paper investigates the recently emerged problem of Language-assisted Image Clustering (LaIC), where textual semantics are leveraged to improve the discriminability of visual representations to facilitate image clustering. Due to the unavailability of true class names, one of core challenges of LaIC lies in how to filter positive nouns, i.e., those semantically close to the images of interest, from unlabeled wild corpus data. Existing filtering strategies are predominantly based on the off-the-shelf feature space learned by CLIP; however, despite being intuitive, these strategies lack a rigorous theoretical foundation. To fill this gap, we propose a novel gradient-based framework, termed as GradNorm, which is theoretically guaranteed and shows strong empirical performance. In particular, we measure the positiveness of each noun based on the magnitude of gradients back-propagated from the cross-entropy between the predicted target distribution and the softmax output. Theoretically, we provide a rigorous error bound to quantify the separability of positive nouns by GradNorm and prove that GradNorm naturally subsumes existing filtering strategies as extremely special cases of itself. Empirically, extensive experiments show that GradNorm achieves the state-of-the-art clustering performance on various benchmarks. Code is publicly available at \href{https://github.com/60pen9/On-the-Provable-Importance-of-Gradients-for-Language-Assisted-Image-Clustering}{here}.
format Preprint
id arxiv_https___arxiv_org_abs_2510_16335
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle On the Provable Importance of Gradients for Language-Assisted Image Clustering
Peng, Bo
Lu, Jie
Zhang, Guangquan
Fang, Zhen
Computer Vision and Pattern Recognition
This paper investigates the recently emerged problem of Language-assisted Image Clustering (LaIC), where textual semantics are leveraged to improve the discriminability of visual representations to facilitate image clustering. Due to the unavailability of true class names, one of core challenges of LaIC lies in how to filter positive nouns, i.e., those semantically close to the images of interest, from unlabeled wild corpus data. Existing filtering strategies are predominantly based on the off-the-shelf feature space learned by CLIP; however, despite being intuitive, these strategies lack a rigorous theoretical foundation. To fill this gap, we propose a novel gradient-based framework, termed as GradNorm, which is theoretically guaranteed and shows strong empirical performance. In particular, we measure the positiveness of each noun based on the magnitude of gradients back-propagated from the cross-entropy between the predicted target distribution and the softmax output. Theoretically, we provide a rigorous error bound to quantify the separability of positive nouns by GradNorm and prove that GradNorm naturally subsumes existing filtering strategies as extremely special cases of itself. Empirically, extensive experiments show that GradNorm achieves the state-of-the-art clustering performance on various benchmarks. Code is publicly available at \href{https://github.com/60pen9/On-the-Provable-Importance-of-Gradients-for-Language-Assisted-Image-Clustering}{here}.
title On the Provable Importance of Gradients for Language-Assisted Image Clustering
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2510.16335