Remodeling Semantic Relationships in Vision-Language Fine-Tuning

• URL: http://arxiv.org/abs/2511.08238v2
• Date: Fri, 14 Nov 2025 01:40:35 GMT
• Title: Remodeling Semantic Relationships in Vision-Language Fine-Tuning
• Authors: Xiangyang Wu, Liu Liu, Baosheng Yu, Jiayan Qiu, Zhenwei Shi,
• Abstract summary: We propose a method that can improve multimodal alignment and fusion based on both semantics and relationships.<n>We learn to project the vision features to group related semantics, among which are more likely to have relationships.<n>Finally, we fuse the visual features with the textual by using inheritable cross-attention, where we globally remove the redundant visual relationships.
• Score: 41.69418068980686
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Vision-language fine-tuning has emerged as an efficient paradigm for constructing multimodal foundation models. While textual context often highlights semantic relationships within an image, existing fine-tuning methods typically overlook this information when aligning vision and language, thus leading to suboptimal performance. Toward solving this problem, we propose a method that can improve multimodal alignment and fusion based on both semantics and relationships.Specifically, we first extract multilevel semantic features from different vision encoder to capture more visual cues of the relationships. Then, we learn to project the vision features to group related semantics, among which are more likely to have relationships. Finally, we fuse the visual features with the textual by using inheritable cross-attention, where we globally remove the redundant visual relationships by discarding visual-language feature pairs with low correlation. We evaluate our proposed method on eight foundation models and two downstream tasks, visual question answering and image captioning, and show that it outperforms all existing methods.

Related papers

• GLAD: Generative Language-Assisted Visual Tracking for Low-Semantic Templates [48.65964582402597]
  Vision-language tracking has gained increasing attention in many scenarios.<n>Current vision-language trackers usually employ Transformer architectures for interactive integration of template, search, and text features.<n>We introduce a pioneering Generative Language-AssisteD tracking model, GLAD, which utilizes diffusion models for the generative multi-modal fusion of text description and template image.
  arXiv  Detail & Related papers  (2026-01-31T07:24:56Z)
• Mind-the-Glitch: Visual Correspondence for Detecting Inconsistencies in Subject-Driven Generation [120.23172120151821]
  We propose a novel approach for disentangling visual and semantic features from the backbones of pre-trained diffusion models.<n>We introduce an automated pipeline that constructs image pairs with annotated semantic and visual correspondences.<n>We propose a new metric, Visual Semantic Matching, that quantifies visual inconsistencies in subject-driven image generation.
  arXiv  Detail & Related papers  (2025-09-26T07:11:55Z)
• Embedding and Enriching Explicit Semantics for Visible-Infrared Person Re-Identification [31.011118085494942]
  Visible-infrared person re-identification (VIReID) retrieves pedestrian images with the same identity across different modalities.<n>Existing methods learn visual content solely from images, lacking the capability to sense high-level semantics.<n>We propose an Embedding and Enriching Explicit Semantics framework to learn semantically rich cross-modality pedestrian representations.
  arXiv  Detail & Related papers  (2024-12-11T14:27:30Z)
• Hire: Hybrid-modal Interaction with Multiple Relational Enhancements for Image-Text Matching [7.7559623054251]
  Image-text matching (ITM) is a fundamental problem in computer vision. We propose a Hybrid-modal feature the Interaction with multiple Enhancements (termed textitHire) for image-text matching. In particular, the explicit intra-modal spatial-semantic graph-based reasoning network is designed to improve the contextual representation of visual objects.
  arXiv  Detail & Related papers  (2024-06-05T13:10:55Z)
• Fuse & Calibrate: A bi-directional Vision-Language Guided Framework for Referring Image Segmentation [8.383431263616105]
  We introduce FCNet, a framework that employs a bi-directional guided fusion approach where both vision and language play guiding roles. Specifically, we use a vision-guided approach to conduct initial multi-modal fusion, obtaining multi-modal features that focus on key vision information. We then propose a language-guided calibration module to further calibrate these multi-modal features, ensuring they understand the context of the input sentence.
  arXiv  Detail & Related papers  (2024-05-18T07:21:12Z)
• Coarse-to-Fine Contrastive Learning in Image-Text-Graph Space for Improved Vision-Language Compositionality [50.48859793121308]
  Contrastively trained vision-language models have achieved remarkable progress in vision and language representation learning. Recent research has highlighted severe limitations in their ability to perform compositional reasoning over objects, attributes, and relations.
  arXiv  Detail & Related papers  (2023-05-23T08:28:38Z)
• Learning to Model Multimodal Semantic Alignment for Story Visualization [58.16484259508973]
  Story visualization aims to generate a sequence of images to narrate each sentence in a multi-sentence story. Current works face the problem of semantic misalignment because of their fixed architecture and diversity of input modalities. We explore the semantic alignment between text and image representations by learning to match their semantic levels in the GAN-based generative model.
  arXiv  Detail & Related papers  (2022-11-14T11:41:44Z)
• Single-Stream Multi-Level Alignment for Vision-Language Pretraining [103.09776737512078]
  We propose a single stream model that aligns the modalities at multiple levels. We achieve this using two novel tasks: symmetric cross-modality reconstruction and a pseudo-labeled key word prediction. We demonstrate top performance on a set of Vision-Language downstream tasks such as zero-shot/fine-tuned image/text retrieval, referring expression, and VQA.
  arXiv  Detail & Related papers  (2022-03-27T21:16:10Z)
• Exploring Semantic Relationships for Unpaired Image Captioning [40.401322131624866]
  We achieve unpaired image captioning by bridging the vision and the language domains with high-level semantic information. We propose the Semantic Relationship Explorer, which explores the relationships between semantic concepts for better understanding of the image. The proposed approach boosts five strong baselines under the paired setting, where the most significant improvement in CIDEr score reaches 8%.
  arXiv  Detail & Related papers  (2021-06-20T09:10:11Z)
• Improving Image Captioning with Better Use of Captions [65.39641077768488]
  We present a novel image captioning architecture to better explore semantics available in captions and leverage that to enhance both image representation and caption generation. Our models first construct caption-guided visual relationship graphs that introduce beneficial inductive bias using weakly supervised multi-instance learning. During generation, the model further incorporates visual relationships using multi-task learning for jointly predicting word and object/predicate tag sequences.
  arXiv  Detail & Related papers  (2020-06-21T14:10:47Z)

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.