paper-with-me

홈 › Papers

Revisit Large-Scale Image-Caption Data in Pre-training Multimodal Foundation Models

2024-10-03 · Zhengfeng Lai, Vasileios Saveris, Chen Chen, Hong-You Chen, Haotian Zhang, BoWen Zhang, Juan Lao Tebar, Wenze Hu, Zhe Gan, Peter Grasch, Meng Cao, Yinfei Yang

Recent advancements in multimodal models highlight the value of rewritten captions for improving performance, yet key challenges remain. For example, while synthetic captions often provide superior quality and image-text alignment, it is not clear whether they can fully replace AltTexts: the role of synthetic captions and their interaction with original web-crawled AltTexts in pre-training is still not well understood. Moreover, different multimodal foundation models may have unique preferences for specific caption formats, but efforts to identify the optimal captions for each model remain limited. In this work, we propose a novel, controllable, and scalable captioning pipeline designed to generate diverse caption formats tailored to various multimodal models. By examining Short Synthetic Captions (SSC) towards Dense Synthetic Captions (DSC+) as case studies, we systematically explore their effects and interactions with AltTexts across models such as CLIP, multimodal LLMs, and diffusion models. Our findings reveal that a hybrid approach that keeps both synthetic captions and AltTexts can outperform the use of synthetic captions alone, improving both alignment and performance, with each model demonstrating preferences for particular caption formats. This comprehensive analysis provides valuable insights into optimizing captioning strategies, thereby advancing the pre-training of multimodal foundation models.

📄 PDF Abstract BibTeX arXiv:2410.02740

Code (1)

apple/ml-veclip tf

Methods 이 논문이 사용한 방법론

CLIP Contrastive Language-Image Pre-training (CLIP), consisting of a simplified version of ConVIRT trained from scratch, is an efficient method of image representation learning…
Diffusion Diffusion models generate samples by gradually removing noise from a signal, and their training objective can be expressed as a reweighted variational lower-bound…

Similar Papers 제목 키워드 기반

Reading Is Believing: Revisiting Language Bottleneck Models for Image Classification

2024-06-22 · Honori Udo, Takafumi Koshinaka

We revisit language bottleneck models as an approach to ensuring the explainability of deep learning models for image classification. Because of inevitable information loss incurred in the step of converting images into …

Classificationimage-classificationImage ClassificationLanguage Modeling+1

PMC-InterCPT: Rethinking Biomedical Interleaved Data for Multimodal Continued Pretraining

2026-05-31 · Guanghao Zhu, Zeyu Liu, Zhitian Hou, Pengkai Wang 외 arxiv

Large-scale biomedical image-text datasets extracted from scientific literature provide valuable resources for medical multimodal model training. These datasets are commonly organized as image-caption pairs; however, fig…

CycleCap: Improving VLMs Captioning Performance via Self-Supervised Cycle Consistency Fine-Tuning

2026-03-18 · Marios Krestenitis, Christos Tzelepis, Konstantinos Ioannidis, Stefanos Vrochidis 외 arxiv

Visual-Language Models (VLMs) have achieved remarkable progress in image captioning, visual question answering, and visual reasoning. Yet they remain prone to vision-language misalignment, often producing overly generic …

Visual Question AnsweringVisual ReasoningImage Captioning

BLUEX Revisited: Enhancing Benchmark Coverage with Automatic Captioning

2025-08-29 · João Guilherme Alves Santos, Giovana Kerche Bonás, Thales Sales Almeida arxiv

With the growing capabilities of Large Language Models (LLMs), there is an increasing need for robust evaluation methods, especially in multilingual and non-English contexts. We present an updated version of the BLUEX da…

Large-Scale Bidirectional Training for Zero-Shot Image Captioning

2022-11-13 · TaeHoon Kim, Mark Marsden, Pyunghwan Ahn, Sangyun Kim 외

When trained on large-scale datasets, image captioning models can understand the content of images from a general domain but often fail to generate accurate, detailed captions. To improve performance, pretraining-and-fin…

Image CaptioningKeyword Extraction