Masked Diffusion Captioning for Visual Feature Learning

• URL: http://arxiv.org/abs/2510.26799v1
• Date: Thu, 30 Oct 2025 17:59:46 GMT
• Title: Masked Diffusion Captioning for Visual Feature Learning
• Authors: Chao Feng, Zihao Wei, Andrew Owens,
• Abstract summary: We learn visual features by captioning images with an image-conditioned masked diffusion language model.<n>After training, the learned visual features can be applied to downstream vision tasks.
• Score: 39.550655996569
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: We learn visual features by captioning images with an image-conditioned masked diffusion language model, a formulation we call masked diffusion captioning (MDC). During training, text tokens in each image-caption pair are masked at a randomly chosen ratio, and a decoder conditioned on visual features is trained to reconstruct the original text. After training, the learned visual features can be applied to downstream vision tasks. Unlike autoregressive captioning, the strength of the visual learning signal in MDC does not depend on each token's position in the sequence, reducing the need for auxiliary objectives. Linear probing experiments across a variety of academic-scale models and datasets show that the learned visual features are competitive with those produced by autoregressive and contrastive approaches.

Related papers

• Instruction Tuning-free Visual Token Complement for Multimodal LLMs [51.138806401996696]
  multimodal large language models (MLLMs) have promised an elegant bridge between vision and language. We propose a Visual Token Complement framework (VTC) that helps MLLMs regain the missing visual features. Our VTC integrates text-to-image generation as a guide to identifying the text-irrelevant features, and a visual selector is then developed to generate complementary visual tokens.
  arXiv  Detail & Related papers  (2024-08-09T12:13:01Z)
• ControlMLLM: Training-Free Visual Prompt Learning for Multimodal Large Language Models [73.34709921061928]
  We propose a training-free method to inject visual prompts into Multimodal Large Language Models (MLLMs)<n>We optimize a learnable latent variable based on an energy function, enhancing the strength of referring regions in the attention map.<n>Our method offers a promising direction for integrating referring abilities into MLLMs, and supports referring with box, mask, scribble and point.
  arXiv  Detail & Related papers  (2024-07-31T11:40:29Z)
• Efficient Vision-Language Pre-training by Cluster Masking [13.845233914223561]
  We propose a simple strategy for masking image patches during visual-language contrastive learning. We randomly mask clusters of visually similar image patches, as measured by their raw pixel intensities. This provides an extra learning signal, beyond the contrastive training itself, since it forces a model to predict words for masked visual structures solely from context.
  arXiv  Detail & Related papers  (2024-05-14T17:59:40Z)
• Multi-Modal Representation Learning with Text-Driven Soft Masks [48.19806080407593]
  We propose a visual-linguistic representation learning approach within a self-supervised learning framework. We generate diverse features for the image-text matching (ITM) task via soft-masking the regions in an image. We identify the relevant regions to each word by computing the word-conditional visual attention using multi-modal encoder.
  arXiv  Detail & Related papers  (2023-04-03T05:07:49Z)
• Unleashing Text-to-Image Diffusion Models for Visual Perception [84.41514649568094]
  VPD (Visual Perception with a pre-trained diffusion model) is a new framework that exploits the semantic information of a pre-trained text-to-image diffusion model in visual perception tasks. We show that VPD can be faster adapted to downstream visual perception tasks using the proposed VPD.
  arXiv  Detail & Related papers  (2023-03-03T18:59:47Z)
• VLMAE: Vision-Language Masked Autoencoder [21.97700040013084]
  We propose a vision-language masked autoencoder framework (VLMAE) for vision-language pre-training. VLMAE employs visual generative learning, facilitating the model to acquire fine-grained and unbiased features.
  arXiv  Detail & Related papers  (2022-08-19T14:39:18Z)
• Masked Vision and Language Modeling for Multi-modal Representation Learning [62.15254888833132]
  We study how to use masked signal modeling in vision and language (V+L) representation learning. We propose to build joint masked vision and language modeling, where the masked signal of one modality is reconstructed with the help from another modality. Our experiments on various V+L tasks show that the proposed method achieves state-of-the-art performances by using a large amount of data.
  arXiv  Detail & Related papers  (2022-08-03T15:11:01Z)
• Retrieval-Augmented Transformer for Image Captioning [51.79146669195357]
  We develop an image captioning approach with a kNN memory, with which knowledge can be retrieved from an external corpus to aid the generation process. Our architecture combines a knowledge retriever based on visual similarities, a differentiable encoder, and a kNN-augmented attention layer to predict tokens. Experimental results, conducted on the COCO dataset, demonstrate that employing an explicit external memory can aid the generation process and increase caption quality.
  arXiv  Detail & Related papers  (2022-07-26T19:35:49Z)
• Visual-aware Attention Dual-stream Decoder for Video Captioning [12.139806877591212]
  The attention mechanism in the current video captioning method learns to assign weight to each frame, promoting the decoder dynamically. This may not explicitly model the correlation and the temporal coherence of the visual features extracted in the sequence frames. We propose a new Visual-aware Attention (VA) model, which unifies changes of temporal sequence frames with the words at the previous moment. The effectiveness of the proposed Visual-aware Attention Dual-stream Decoder (VADD) is demonstrated.
  arXiv  Detail & Related papers  (2021-10-16T14:08:20Z)
• Learning Visual Representations with Caption Annotations [19.24013129952071]
  We propose a proxy task to learn visual representations over image-caption pairs. ICMLM consists in predicting masked words in captions by relying on visual cues. Our experiments confirm that image captions can be leveraged to inject global and localized semantic information into visual representations.
  arXiv  Detail & Related papers  (2020-08-04T08:04:16Z)

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.