UniCVR: From Alignment to Reranking for Unified Zero-Shot Composed Visual Retrieval

TL;DR

UniCVR is a unified zero-shot framework that combines multimodal language models and vision-language pre-trained models to perform composed visual retrieval across images and videos without task-specific data.

Contribution

It introduces a novel two-stage approach with contrastive learning and dual-level reranking, unifying three retrieval tasks in a zero-shot setting.

Findings

Achieves state-of-the-art results on five benchmarks.

Effectively unifies image and video retrieval tasks.

Demonstrates strong generalization without task-specific annotations.

Abstract

Composed image retrieval, multi-turn composed image retrieval, and composed video retrieval all share a common paradigm: composing the reference visual with modification text to retrieve the desired target. Despite this shared structure, the three tasks have been studied in isolation, with no prior work proposing a unified framework, let alone a zero-shot solution. In this paper, we propose UniCVR, the first unified zero-shot composed visual retrieval framework that jointly addresses all three tasks without any task-specific human-annotated data. UniCVR strategically combines two complementary strengths: Multimodal Large Language Models (MLLMs) for compositional query understanding and Vision-Language Pre-trained (VLP) models for structured visual retrieval. Concretely, UniCVR operates in two stages. In Stage I, we train the MLLM as a compositional query embedder via contrastive learning on a curated multi-source dataset of approximately 3.5M samples, bridging the heterogeneous embedding spaces between the MLLM and the frozen VLP gallery encoder. A cluster-based hard negative sampling strategy is proposed to strengthen contrastive supervision. In Stage II, we introduce an MLLM-guided dual-level reranking mechanism that applies adaptive budgeted subset scoring to a small number of top-ranked candidates, and then exploits the resulting relevance signals through a dual-level re-scoring scheme, producing more accurate final rankings with minimal computational overhead. Extensive experiments across five benchmarks covering all three tasks demonstrate that UniCVR achieves cutting-edge performance, validating its effectiveness and generalizability. Our data and code will be released upon acceptance.