# Self-Training Boosted Multi-Factor Matching Network for Composed Image Retrieval

arXiv: https://arxiv.org/abs/2305.09979 · HTML: https://mlsanity.com/paper/2305.09979/

## Key facts

- Cited by 22 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 17 May 2023.

## Built on by (most cited first, 22 of 22)

- [FineCIR: Explicit Parsing of Fine-Grained Modification Semantics for Composed Image Retrieval](https://mlsanity.com/paper/2503.21309/) (2025)
- [OFFSET: Segmentation-based Focus Shift Revision for Composed Image Retrieval](https://mlsanity.com/paper/2507.05631/) (2025)
- [HUD: Hierarchical Uncertainty-Aware Disambiguation Network for Composed Video Retrieval](https://mlsanity.com/paper/2512.02792/) (2025)
- [Fine-grained Textual Inversion Network for Zero-Shot Composed Image Retrieval](https://mlsanity.com/paper/2503.19296/) (2025)
- [A Comprehensive Survey on Composed Image Retrieval](https://mlsanity.com/paper/2502.18495/) (2025)
- [VINCIE: Unlocking In-context Image Editing from Video](https://mlsanity.com/paper/2506.10941/) (2025)
- [Visual Delta Generator with Large Multi-modal Models for Semi-supervised Composed Image Retrieval](https://mlsanity.com/paper/2404.15516/) (2024)
- [COMBINER: Composed Image Retrieval Guided by Attribute-based Neighbor Relations](https://mlsanity.com/paper/2606.04604/) (2026)
- [Multimodal Reasoning Agent for Zero-Shot Composed Image Retrieval](https://mlsanity.com/paper/2505.19952/) (2025)
- [MLLM-Guided VLM Fine-Tuning with Joint Inference for Zero-Shot Composed Image Retrieval](https://mlsanity.com/paper/2505.19707/) (2025)
- [Pseudo-triplet Guided Few-shot Composed Image Retrieval](https://mlsanity.com/paper/2407.06001/) (2024)
- [TTOM: Test-Time Optimization and Memorization for Compositional Video Generation](https://mlsanity.com/paper/2510.07940/) (2025)
- [Self-distilled Dynamic Fusion Network for Language-based Fashion Retrieval](https://mlsanity.com/paper/2405.15451/) (2024)
- [CalibCLIP: Contextual Calibration of Dominant Semantics for Text-Driven Image Retrieval](https://mlsanity.com/paper/2510.05586/) (2025)
- [ReCALL: Recalibrating Capability Degradation for MLLM-based Composed Image Retrieval](https://mlsanity.com/paper/2602.01639/) (2026)
- [Composed Multi-modal Retrieval: A Survey of Approaches and Applications](https://mlsanity.com/paper/2503.01334/) (2025)
- [Data-Efficient Generalization for Zero-shot Composed Image Retrieval](https://mlsanity.com/paper/2503.05204/) (2025)
- [UniCVR: From Alignment to Reranking for Unified Zero-Shot Composed Visual Retrieval](https://mlsanity.com/paper/2604.20318/) (2026)
- [Fine-Grained Zero-Shot Composed Image Retrieval with Complementary Visual-Semantic Integration](https://mlsanity.com/paper/2601.14060/) (2026)
- [Never Seen Before: Benchmarking Genuine Zero-Shot Composed Image Retrieval with Consistent Video-Sourced Datasets](https://mlsanity.com/paper/2606.07032/) (2026)
- [Composed Object Retrieval: Object-level Retrieval via Composed Expressions](https://mlsanity.com/paper/2508.04424/) (2025)
- [Thinking Before Retrieving: Robust Zero-Shot Composed Image Retrieval via Strategic Planning and Self-Criticism](https://mlsanity.com/paper/2606.31222/) (2026)

## Abstract (arXiv)

The composed image retrieval (CIR) task aims to retrieve the desired target image for a given multimodal query, i.e., a reference image with its corresponding modification text. The key limitations encountered by existing efforts are two aspects: 1) ignoring the multi-faceted query-target matching factors; 2) ignoring the potential unlabeled reference-target image pairs in existing benchmark datasets. To address these two limitations is non-trivial due to the following challenges: 1) how to effectively model the multi-faceted matching factors in a latent way without direct supervision signals; 2) how to fully utilize the potential unlabeled reference-target image pairs to improve the generalization ability of the CIR model. To address these challenges, in this work, we first propose a muLtI-faceted Matching Network (LIMN), which consists of three key modules: multi-grained image/text encoder, latent factor-oriented feature aggregation, and query-target matching modeling. Thereafter, we design an iterative dual self-training paradigm to further enhance the performance of LIMN by fully utilizing the potential unlabeled reference-target image pairs in a semi-supervised manner. Specifically, we denote the iterative dual self-training paradigm enhanced LIMN as LIMN+. Extensive experiments on three real-world datasets, FashionIQ, Shoes, and Birds-to-Words, show that our proposed method significantly surpasses the state-of-the-art baselines.