# FAR-Net: Multi-Stage Fusion Network with Enhanced Semantic Alignment and Adaptive Reconciliation for Composed Image Retrieval

arXiv: https://arxiv.org/abs/2507.12823 · HTML: https://mlsanity.com/paper/2507.12823/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Builds on 19 of its 34 references that are in the map.
- First posted on arXiv 17 Jul 2025.

## Builds on (most cited first, 19 of 19)

- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [BLIP-2: Bootstrapping Language-Image Pre-training with Frozen Image Encoders and Large Language Models](https://mlsanity.com/paper/2301.12597/) (2023)
- [CLIP4Clip: An Empirical Study of CLIP for End to End Video Clip Retrieval](https://mlsanity.com/paper/2104.08860/) (2021)
- [Image Retrieval on Real-life Images with Pre-trained Vision-and-Language Models](https://mlsanity.com/paper/2108.04024/) (2021)
- [Composing Text and Image for Image Retrieval - An Empirical Odyssey](https://mlsanity.com/paper/1812.07119/) (2018)
- [Empowering Things with Intelligence: A Survey of the Progress, Challenges, and Opportunities in Artificial Intelligence of Things](https://mlsanity.com/paper/2011.08612/) (2020)
- [CompoDiff: Versatile Composed Image Retrieval With Latent Diffusion](https://mlsanity.com/paper/2303.11916/) (2023)
- [ARTEMIS: Attention-based Retrieval with Text-Explicit Matching and Implicit Similarity](https://mlsanity.com/paper/2203.08101/) (2022)
- [Sentence-level Prompts Benefit Composed Image Retrieval](https://mlsanity.com/paper/2310.05473/) (2023)
- [Data Roaming and Quality Assessment for Composed Image Retrieval](https://mlsanity.com/paper/2303.09429/) (2023)
- [Composed Image Retrieval with Text Feedback via Multi-grained Uncertainty Regularization](https://mlsanity.com/paper/2211.07394/) (2022)
- [Target-Guided Composed Image Retrieval](https://mlsanity.com/paper/2309.01366/) (2023)
- [MMFT-BERT: Multimodal Fusion Transformer with BERT Encodings for Visual Question Answering](https://mlsanity.com/paper/2010.14095/) (2020)
- [Simple but Effective Raw-Data Level Multimodal Fusion for Composed Image Retrieval](https://mlsanity.com/paper/2404.15875/) (2024)
- [Dynamic Weighted Combiner for Mixed-Modal Image Retrieval](https://mlsanity.com/paper/2312.06179/) (2023)
- [SyncMask: Synchronized Attentional Masking for Fashion-centric Vision-Language Pretraining](https://mlsanity.com/paper/2404.01156/) (2024)
- [Multimodal Inverse Cloze Task for Knowledge-based Visual Question Answering](https://mlsanity.com/paper/2301.04366/) (2023)

## Built on by (most cited first, 3 of 3)

- [DAFM: Dynamic Adaptive Fusion for Multi-Model Collaboration in Composed Image Retrieval](https://mlsanity.com/paper/2511.05020/) (2025)
- [Thinking Before Retrieving: Robust Zero-Shot Composed Image Retrieval via Strategic Planning and Self-Criticism](https://mlsanity.com/paper/2606.31222/) (2026)
- [DiCE-CIR: Direct Composition Learning for Efficient Zero-Shot Composed Image Retrieval](https://mlsanity.com/paper/2607.04665/) (2026)

## Abstract (arXiv)

Composed image retrieval (CIR) is a vision language task that retrieves a target image using a reference image and modification text, enabling intuitive specification of desired changes. While effectively fusing visual and textual modalities is crucial, existing methods typically adopt either early or late fusion. Early fusion tends to excessively focus on explicitly mentioned textual details and neglect visual context, whereas late fusion struggles to capture fine-grained semantic alignments between image regions and textual tokens. To address these issues, we propose FAR-Net, a multi-stage fusion framework designed with enhanced semantic alignment and adaptive reconciliation, integrating two complementary modules. The enhanced semantic alignment module (ESAM) employs late fusion with cross-attention to capture fine-grained semantic relationships, while the adaptive reconciliation module (ARM) applies early fusion with uncertainty embeddings to enhance robustness and adaptability. Experiments on CIRR and FashionIQ show consistent performance gains, improving Recall@1 by up to 2.4% and Recall@50 by 1.04% over existing state-of-the-art methods, empirically demonstrating that FAR Net provides a robust and scalable solution to CIR tasks.