# Rethinking Text-Based Image Retrieval in Specific Domain

arXiv: https://arxiv.org/abs/2608.10524 · HTML: https://mlsanity.com/paper/2608.10524/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 24 of its 31 references that are in the map.
- First posted on arXiv 11 Aug 2026.

## Builds on (most cited first, 24 of 24)

- [Microsoft COCO: Common Objects in Context](https://mlsanity.com/paper/1405.0312/) (2014)
- [DINOv2: Learning Robust Visual Features without Supervision](https://mlsanity.com/paper/2304.07193/) (2023)
- [Scaling Up Visual and Vision-Language Representation Learning With Noisy Text Supervision](https://mlsanity.com/paper/2102.05918/) (2021)
- [Visual Genome: Connecting Language and Vision Using Crowdsourced Dense Image Annotations](https://mlsanity.com/paper/1602.07332/) (2016)
- [LAION-400M: Open Dataset of CLIP-Filtered 400 Million Image-Text Pairs](https://mlsanity.com/paper/2111.02114/) (2021)
- [Flickr30k Entities: Collecting Region-to-Phrase Correspondences for Richer Image-to-Sentence Models](https://mlsanity.com/paper/1505.04870/) (2015)
- [Reproducible scaling laws for contrastive language-image learning](https://mlsanity.com/paper/2212.07143/) (2022)
- [When and why vision-language models behave like bags-of-words, and what to do about it?](https://mlsanity.com/paper/2210.01936/) (2022)
- [Qwen3-VL-Embedding and Qwen3-VL-Reranker: A Unified Framework for State-of-the-Art Multimodal Retrieval and Ranking](https://mlsanity.com/paper/2601.04720/) (2026)
- [Person Search with Natural Language Description](https://mlsanity.com/paper/1702.05729/) (2017)
- [Automatic Spatially-aware Fashion Concept Discovery](https://mlsanity.com/paper/1708.01311/) (2017)
- [Semantically Self-Aligned Network for Text-to-Image Part-aware Person Re-identification](https://mlsanity.com/paper/2107.12666/) (2021)
- [DSSL: Deep Surroundings-person Separation Learning for Text-based Person Retrieval](https://mlsanity.com/paper/2109.05534/) (2021)
- [TinyCLIP: CLIP Distillation via Affinity Mimicking and Weight Inheritance](https://mlsanity.com/paper/2309.12314/) (2023)
- [Breaking the Modality Barrier: Universal Embedding Learning with Multimodal LLMs](https://mlsanity.com/paper/2504.17432/) (2025)
- [Detect Anything via Next Point Prediction](https://mlsanity.com/paper/2510.12798/) (2025)
- [SoftCLIP: Softer Cross-modal Alignment Makes CLIP Stronger](https://mlsanity.com/paper/2303.17561/) (2023)
- [PLIP: Language-Image Pre-training for Person Representation Learning](https://mlsanity.com/paper/2305.08386/) (2023)
- [UniME-V2: MLLM-as-a-Judge for Universal Multimodal Embedding Learning](https://mlsanity.com/paper/2510.13515/) (2025)
- [INQUIRE: A Natural World Text-to-Image Retrieval Benchmark](https://mlsanity.com/paper/2411.02537/) (2024)
- [CWCL: Cross-Modal Transfer with Continuously Weighted Contrastive Loss](https://mlsanity.com/paper/2309.14580/) (2023)
- [e-CLIP: Large-Scale Vision-Language Representation Learning in E-commerce](https://mlsanity.com/paper/2207.00208/) (2022)
- [Bringing CLIP to the Clinic: Dynamic Soft Labels and Negation-Aware Learning for Medical Analysis](https://mlsanity.com/paper/2505.22079/) (2025)
- [CellCLIP -- Learning Perturbation Effects in Cell Painting via Text-Guided Contrastive Learning](https://mlsanity.com/paper/2506.06290/) (2025)

## Abstract (arXiv)

Driven by the rapid advancement of vision-language representation learning, Text-based Image Retrieval (TBIR) has made notable progress. However, existing benchmarks are predominantly constructed on an exclusive single-match assumption between query and images. While effective in general scenarios, this assumption fails to reflect practical system performance in specific domains (e.g., surveillance), where a single query often corresponds to multiple relevant candidate images. To address this limitation, we design a Domain-Specific Multi-Match Text-based Image Retrieval (DSMM-TBIR) data engine. Leveraging this engine, we construct Security Multi-Match TBIR (SecMM-TBIR), a benchmark comprising 50k surveillance images with 200 comprehensive queries. Furthermore, we observe that vanilla contrastive learning in specific domains suffers from severe false negatives, forcing the model to push apart semantically similar pairs and thus degrading retrieval performance. We propose the Semantic-Aware Fine-Tuning (SAFT) framework to address semantic compression in specific domains, which incorporates Semantic-Aware Soft-Label Supervision (SASS) and Intra-modal Structural Distillation (ISD) to establish a promising paradigm for domain-specific TBIR tasks. Experiments across diverse CLIP-like models demonstrate that SAFT yields an average mAP@20 gain of 7.8 points on SecMM-TBIR over standard image-text contrastive (ITC) fine-tuning, while also improving general-domain performance. The entire benchmark will be released to facilitate further research.