Vision-Free Retrieval: Rethinking Multimodal Search with Textual Scene Descriptions

• URL: http://arxiv.org/abs/2509.19203v1
• Date: Tue, 23 Sep 2025 16:22:27 GMT
• Title: Vision-Free Retrieval: Rethinking Multimodal Search with Textual Scene Descriptions
• Authors: Ioanna Ntinou, Alexandros Xenos, Yassine Ouali, Adrian Bulat, Georgios Tzimiropoulos,
• Abstract summary: We introduce a vision-free, single-encoder retrieval pipeline for vision-language models.<n>We migrate to a text-to-text paradigm with the assistance of VLLM-generated structured image descriptions.<n>Our approach achieves state-of-the-art zero-shot performance on multiple retrieval and compositionality benchmarks.
• Score: 81.33113485830711
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Contrastively-trained Vision-Language Models (VLMs), such as CLIP, have become the standard approach for learning discriminative vision-language representations. However, these models often exhibit shallow language understanding, manifesting bag-of-words behaviour. These limitations are reinforced by their dual-encoder design, which induces a modality gap. Additionally, the reliance on vast web-collected data corpora for training makes the process computationally expensive and introduces significant privacy concerns. To address these limitations, in this work, we challenge the necessity of vision encoders for retrieval tasks by introducing a vision-free, single-encoder retrieval pipeline. Departing from the traditional text-to-image retrieval paradigm, we migrate to a text-to-text paradigm with the assistance of VLLM-generated structured image descriptions. We demonstrate that this paradigm shift has significant advantages, including a substantial reduction of the modality gap, improved compositionality, and better performance on short and long caption queries, all attainable with only a few hours of calibration on two GPUs. Additionally, substituting raw images with textual descriptions introduces a more privacy-friendly alternative for retrieval. To further assess generalisation and address some of the shortcomings of prior compositionality benchmarks, we release two benchmarks derived from Flickr30k and COCO, containing diverse compositional queries made of short captions, which we coin subFlickr and subCOCO. Our vision-free retriever matches and often surpasses traditional multimodal models. Importantly, our approach achieves state-of-the-art zero-shot performance on multiple retrieval and compositionality benchmarks, with models as small as 0.3B parameters. Code is available at: https://github.com/IoannaNti/LexiCLIP

Related papers

• When Vision Meets Texts in Listwise Reranking [1.2691047660244335]
  Rank-Nexus is a multimodal image-text document reranker that performs listwise qualitative reranking on retrieved lists incorporating both images and texts.<n>We first train modalities separately: leveraging abundant text reranking data, we distill knowledge into the text branch.<n>For images, where data is scarce, we construct distilled pairs from multimodal large language model (MLLM) captions on image retrieval benchmarks.
  arXiv  Detail & Related papers  (2026-01-28T13:57:14Z)
• ViTCoP: Accelerating Large Vision-Language Models via Visual and Textual Semantic Collaborative Pruning [8.933549837045932]
  Large Vision-Language Models incur high computational costs due to significant redundancy in their visual tokens.<n>We propose a Visual and Textual Semantic Collaborative Pruning framework (ViTCoP) that combines redundancy filtering in the vision encoder with step-wise co-pruning within the Large Language Models.
  arXiv  Detail & Related papers  (2026-01-25T12:47:30Z)
• Exploring a Unified Vision-Centric Contrastive Alternatives on Multi-Modal Web Documents [99.62178668680578]
  We propose Vision-Centric Contrastive Learning (VC2L), a unified framework that models text, images, and their combinations using a single vision transformer.<n> VC2L operates entirely in pixel space by rendering all inputs, whether textual, visual, or combined, as images.<n>To capture complex cross-modal relationships in web documents, VC2L employs a snippet-level contrastive learning objective that aligns consecutive multimodal segments.
  arXiv  Detail & Related papers  (2025-10-21T14:59:29Z)
• Generalized Contrastive Learning for Universal Multimodal Retrieval [53.70202081784898]
  Cross-modal retrieval models (e.g., CLIP) show degraded performances with retrieving keys composed of fused image-text modality.<n>This paper proposes Generalized Contrastive Learning (GCL), a novel loss formulation that improves multimodal retrieval performance without the need for new dataset curation.
  arXiv  Detail & Related papers  (2025-09-30T01:25:04Z)
• CoLLM: A Large Language Model for Composed Image Retrieval [76.29725148964368]
  Composed Image Retrieval (CIR) is a complex task that aims to retrieve images based on a multimodal query.<n>We present CoLLM, a one-stop framework that generates triplets on-the-fly from image-caption pairs.<n>We leverage Large Language Models (LLMs) to generate joint embeddings of reference images and modification texts.
  arXiv  Detail & Related papers  (2025-03-25T17:59:50Z)
• A Chain-of-Thought Subspace Meta-Learning for Few-shot Image Captioning with Large Vision and Language Models [17.144311122664508]
  A large-scale vision and language model that has been pretrained on massive data encodes visual and linguistic prior.<n>We propose a chain-of-thought (CoT) meta-learning scheme as a multi-step image captioning procedure to better imitate how humans describe images.
  arXiv  Detail & Related papers  (2025-02-19T18:35:43Z)
• Towards Text-Image Interleaved Retrieval [49.96332254241075]
  We introduce the text-image interleaved retrieval (TIIR) task, where the query and document are interleaved text-image sequences.<n>We construct a TIIR benchmark based on naturally interleaved wikiHow tutorials, where a specific pipeline is designed to generate interleaved queries.<n>We propose a novel Matryoshka Multimodal Embedder (MME), which compresses the number of visual tokens at different granularity.
  arXiv  Detail & Related papers  (2025-02-18T12:00:47Z)
• ViLa-MIL: Dual-scale Vision-Language Multiple Instance Learning for Whole Slide Image Classification [52.405499816861635]
  Multiple instance learning (MIL)-based framework has become the mainstream for processing the whole slide image (WSI)<n>We propose a dual-scale vision-language multiple instance learning (ViLa-MIL) framework for whole slide image classification.
  arXiv  Detail & Related papers  (2025-02-12T13:28:46Z)
• Advancing Myopia To Holism: Fully Contrastive Language-Image Pre-training [30.071860810401933]
  This paper advances contrastive language-image pre-training (CLIP) into one novel holistic paradigm.<n>We use image-to-text captioning to generate multi-texts for each image, from multiple perspectives, granularities, and hierarchies.<n>Our holistic CLIP significantly outperforms existing CLIP, including image-text retrieval, open-vocabulary classification, and dense visual tasks.
  arXiv  Detail & Related papers  (2024-11-30T11:27:58Z)
• MVAM: Multi-View Attention Method for Fine-grained Image-Text Matching [65.87255122130188]
  We propose a Multi-view Attention Method (MVAM) for image-text matching.<n>We also incorporate an objective to explicitly encourage attention heads to focus on distinct aspects of the input data.<n>Our method allows models to encode images and text from different perspectives and focus on more critical details, leading to better matching performance.
  arXiv  Detail & Related papers  (2024-02-27T06:11:54Z)
• Fine-tuning CLIP Text Encoders with Two-step Paraphrasing [83.3736789315201]
  We introduce a straightforward fine-tuning approach to enhance the representations of CLIP models for paraphrases. Our model, which we call ParaCLIP, exhibits significant improvements over baseline CLIP models across various tasks.
  arXiv  Detail & Related papers  (2024-02-23T06:11:50Z)
• LightCLIP: Learning Multi-Level Interaction for Lightweight Vision-Language Models [45.672539931681065]
  We propose a multi-level interaction paradigm for training lightweight CLIP models. An auxiliary fusion module injecting unmasked image embedding into masked text embedding is proposed.
  arXiv  Detail & Related papers  (2023-12-01T15:54:55Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.