Decouple before Align: Visual Disentanglement Enhances Prompt Tuning

• URL: http://arxiv.org/abs/2508.00395v1
• Date: Fri, 01 Aug 2025 07:46:00 GMT
• Title: Decouple before Align: Visual Disentanglement Enhances Prompt Tuning
• Abstract summary: Prompt tuning (PT) has showcased remarkable effectiveness in improving the task-specific transferability of vision-language models.<n>This paper delves into a previously overlooked information asymmetry issue in PT, where the visual modality mostly conveys more context.<n>We propose DAPT, an effective PT framework based on an intuitive decouple-before-align concept.
• Score: 85.91474962071452
• License: http://creativecommons.org/licenses/by-nc-sa/4.0/
• Abstract: Prompt tuning (PT), as an emerging resource-efficient fine-tuning paradigm, has showcased remarkable effectiveness in improving the task-specific transferability of vision-language models. This paper delves into a previously overlooked information asymmetry issue in PT, where the visual modality mostly conveys more context than the object-oriented textual modality. Correspondingly, coarsely aligning these two modalities could result in the biased attention, driving the model to merely focus on the context area. To address this, we propose DAPT, an effective PT framework based on an intuitive decouple-before-align concept. First, we propose to explicitly decouple the visual modality into the foreground and background representation via exploiting coarse-and-fine visual segmenting cues, and then both of these decoupled patterns are aligned with the original foreground texts and the hand-crafted background classes, thereby symmetrically strengthening the modal alignment. To further enhance the visual concentration, we propose a visual pull-push regularization tailored for the foreground-background patterns, directing the original visual representation towards unbiased attention on the region-of-interest object. We demonstrate the power of architecture-free DAPT through few-shot learning, base-to-novel generalization, and data-efficient learning, all of which yield superior performance across prevailing benchmarks. Our code will be released at https://github.com/Ferenas/DAPT.

Related papers

• PRIMED: Adaptive Modality Suppression for Referring Audio-Visual Segmentation via Biased Competition [9.145417637255651]
  Referring Audio-Visual (Ref-AVS) seeks to localize and segment target objects in video frames based on visual, auditory, and textual referring cues.<n>We propose PRIMED, inspired by the biased competition theory in cognitive neuroscience, which explicitly models both visual perception and language-driven prior modulation.
  arXiv  Detail & Related papers  (2026-05-08T02:40:34Z)
• Boosting Point-supervised Temporal Action Localization via Text Refinement and Alignment [66.80402022104074]
  We propose a Text Refinement and Alignment (TRA) framework that effectively utilizes textual features from visual descriptions to complement the visual features as they are semantically rich.<n>This is achieved by designing two new modules for the original point-supervised framework: a Point-based Text Refinement module (PTR) and a Point-based Multimodal Alignment module (PMA)
  arXiv  Detail & Related papers  (2026-02-01T14:35:46Z)
• Towards Mitigating Hallucinations in Large Vision-Language Models by Refining Textual Embeddings [39.83857755243316]
  We identify an inherent bias in prevailing LVLM architectures toward the language modality.<n>We propose a method that refines textual embeddings by integrating average-pooled visual features.<n>Our approach demonstrably improves visual grounding and significantly reduces hallucinations on established benchmarks.
  arXiv  Detail & Related papers  (2025-11-07T06:39:54Z)
• 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)
• Background Matters Too: A Language-Enhanced Adversarial Framework for Person Re-Identification [1.409283414986451]
  We argue that background semantics are as important as the foreground semantics in ReID.<n>This paper proposes an end-to-end framework that jointly models foreground and background information.
  arXiv  Detail & Related papers  (2025-09-03T05:38:22Z)
• Infusing fine-grained visual knowledge to Vision-Language Models [5.487134463783365]
  Large-scale contrastive pre-training produces powerful Vision-and-Language Models (VLMs)<n>We propose a fine-tuning method explicitly designed to achieve optimal balance between fine-grained domain adaptation and retention of the pretrained VLM's broad multimodal knowledge.<n>Our approach consistently achieves strong results, notably retaining the visual-text alignment without utilizing any text data or the original text encoder during fine-tuning.
  arXiv  Detail & Related papers  (2025-08-16T19:12:09Z)
• Top-Down Compression: Revisit Efficient Vision Token Projection for Visual Instruction Tuning [70.57180215148125]
  Visual instruction tuning aims to enable large language models to comprehend the visual world.<n>Existing methods often grapple with the intractable trade-off between accuracy and efficiency.<n>We present LLaVA-Meteor, a novel approach that strategically compresses visual tokens without compromising core information.
  arXiv  Detail & Related papers  (2025-05-17T10:22:29Z)
• Re-Align: Aligning Vision Language Models via Retrieval-Augmented Direct Preference Optimization [18.855378039713678]
  Large Vision Language Models (VLMs) are prone to significant hallucinations, particularly in the form of cross-modal inconsistencies.<n>We introduce Re-Align, a novel alignment framework that leverages image retrieval to construct a dual-preference dataset.<n>We also introduce rDPO, an extension of the standard direct preference optimization that incorporates an additional visual preference objective during fine-tuning.
  arXiv  Detail & Related papers  (2025-02-18T18:59:57Z)
• VeCAF: Vision-language Collaborative Active Finetuning with Training Objective Awareness [56.87603097348203]
  VeCAF uses labels and natural language annotations to perform parametric data selection for PVM finetuning. VeCAF incorporates the finetuning objective to select significant data points that effectively guide the PVM towards faster convergence. On ImageNet, VeCAF uses up to 3.3x less training batches to reach the target performance compared to full finetuning.
  arXiv  Detail & Related papers  (2024-01-15T17:28:37Z)
• Bridging Modality Gap for Visual Grounding with Effecitve Cross-modal Distillation [2.104191333263349]
  Current visual grounding methods leverage pre-trained visual and language backbones independently to obtain visual features and linguistic features. This problem arises from the domain gap between the single-modal pre-training backbones used in current visual grounding methods. We propose an Empowering Pre-trained Model for Visual Grounding framework, which distills a multimodal pre-trained model to guide the visual grounding task.
  arXiv  Detail & Related papers  (2023-12-29T15:32:11Z)
• Harnessing Diffusion Models for Visual Perception with Meta Prompts [68.78938846041767]
  We propose a simple yet effective scheme to harness a diffusion model for visual perception tasks. We introduce learnable embeddings (meta prompts) to the pre-trained diffusion models to extract proper features for perception. Our approach achieves new performance records in depth estimation tasks on NYU depth V2 and KITTI, and in semantic segmentation task on CityScapes.
  arXiv  Detail & Related papers  (2023-12-22T14:40:55Z)
• Vision-Enhanced Semantic Entity Recognition in Document Images via Visually-Asymmetric Consistency Learning [19.28860833813788]
  Existing models commonly train a visual encoder with weak cross-modal supervision signals. We propose a novel textbfVisually-textbfAsymmetric cotextbfNsistentextbfCy textbfLearning (textscVancl) approach to capture fine-grained visual and layout features.
  arXiv  Detail & Related papers  (2023-10-23T10:37:22Z)
• VS-TransGRU: A Novel Transformer-GRU-based Framework Enhanced by Visual-Semantic Fusion for Egocentric Action Anticipation [33.41226268323332]
  Egocentric action anticipation is a challenging task that aims to make advanced predictions of future actions in the first-person view. Most existing methods focus on improving the model architecture and loss function based on the visual input and recurrent neural network. We propose a novel visual-semantic fusion enhanced and Transformer GRU-based action anticipation framework.
  arXiv  Detail & Related papers  (2023-07-08T06:49:54Z)
• Fine-Grained Semantically Aligned Vision-Language Pre-Training [151.7372197904064]
  Large-scale vision-language pre-training has shown impressive advances in a wide range of downstream tasks. Existing methods mainly model the cross-modal alignment by the similarity of the global representations of images and texts. We introduce LO, a fine-grained semantically aLigned visiOn-langUage PrE-training framework, which learns fine-grained semantic alignment from the novel perspective of game-theoretic interactions.
  arXiv  Detail & Related papers  (2022-08-04T07:51:48Z)
• Deep Semantic Matching with Foreground Detection and Cycle-Consistency [103.22976097225457]
  We address weakly supervised semantic matching based on a deep network. We explicitly estimate the foreground regions to suppress the effect of background clutter. We develop cycle-consistent losses to enforce the predicted transformations across multiple images to be geometrically plausible and consistent.
  arXiv  Detail & Related papers  (2020-03-31T22:38:09Z)

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.