Saved in:
Bibliographic Details
Main Authors: Tang, Haomiao, Wang, Jinpeng, Zhao, Minyi, Meng, Guanghao, Luo, Ruisheng, Chen, Long, Xia, Shu-Tao
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2601.11393
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914271827329024
author Tang, Haomiao
Wang, Jinpeng
Zhao, Minyi
Meng, Guanghao
Luo, Ruisheng
Chen, Long
Xia, Shu-Tao
author_facet Tang, Haomiao
Wang, Jinpeng
Zhao, Minyi
Meng, Guanghao
Luo, Ruisheng
Chen, Long
Xia, Shu-Tao
contents Composed Image Retrieval (CIR) enables image search by combining a reference image with modification text. Intrinsic noise in CIR triplets incurs intrinsic uncertainty and threatens the model's robustness. Probabilistic learning approaches have shown promise in addressing such issues; however, they fall short for CIR due to their instance-level holistic modeling and homogeneous treatment of queries and targets. This paper introduces a Heterogeneous Uncertainty-Guided (HUG) paradigm to overcome these limitations. HUG utilizes a fine-grained probabilistic learning framework, where queries and targets are represented by Gaussian embeddings that capture detailed concepts and uncertainties. We customize heterogeneous uncertainty estimations for multi-modal queries and uni-modal targets. Given a query, we capture uncertainties not only regarding uni-modal content quality but also multi-modal coordination, followed by a provable dynamic weighting mechanism to derive comprehensive query uncertainty. We further design uncertainty-guided objectives, including query-target holistic contrast and fine-grained contrasts with comprehensive negative sampling strategies, which effectively enhance discriminative learning. Experiments on benchmarks demonstrate HUG's effectiveness beyond state-of-the-art baselines, with faithful analysis justifying the technical contributions.
format Preprint
id arxiv_https___arxiv_org_abs_2601_11393
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Heterogeneous Uncertainty-Guided Composed Image Retrieval with Fine-Grained Probabilistic Learning
Tang, Haomiao
Wang, Jinpeng
Zhao, Minyi
Meng, Guanghao
Luo, Ruisheng
Chen, Long
Xia, Shu-Tao
Computer Vision and Pattern Recognition
Composed Image Retrieval (CIR) enables image search by combining a reference image with modification text. Intrinsic noise in CIR triplets incurs intrinsic uncertainty and threatens the model's robustness. Probabilistic learning approaches have shown promise in addressing such issues; however, they fall short for CIR due to their instance-level holistic modeling and homogeneous treatment of queries and targets. This paper introduces a Heterogeneous Uncertainty-Guided (HUG) paradigm to overcome these limitations. HUG utilizes a fine-grained probabilistic learning framework, where queries and targets are represented by Gaussian embeddings that capture detailed concepts and uncertainties. We customize heterogeneous uncertainty estimations for multi-modal queries and uni-modal targets. Given a query, we capture uncertainties not only regarding uni-modal content quality but also multi-modal coordination, followed by a provable dynamic weighting mechanism to derive comprehensive query uncertainty. We further design uncertainty-guided objectives, including query-target holistic contrast and fine-grained contrasts with comprehensive negative sampling strategies, which effectively enhance discriminative learning. Experiments on benchmarks demonstrate HUG's effectiveness beyond state-of-the-art baselines, with faithful analysis justifying the technical contributions.
title Heterogeneous Uncertainty-Guided Composed Image Retrieval with Fine-Grained Probabilistic Learning
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2601.11393