paper-with-me

Papers

Bootstrapping Vision-Language Learning with Decoupled Language Pre-training

2023-07-13 · NeurIPS 2023 11 · Yiren Jian, Chongyang Gao, Soroush Vosoughi

We present a novel methodology aimed at optimizing the application of frozen large language models (LLMs) for resource-intensive vision-language (VL) pre-training. The current paradigm uses visual features as prompts to guide language models, with a focus on determining the most relevant visual features for corresponding text. Our approach diverges by concentrating on the language component, specifically identifying the optimal prompts to align with visual features. We introduce the Prompt-Transformer (P-Former), a model that predicts these ideal prompts, which is trained exclusively on linguistic data, bypassing the need for image-text pairings. This strategy subtly bifurcates the end-to-end VL training process into an additional, separate stage. Our experiments reveal that our framework significantly enhances the performance of a robust image-to-text baseline (BLIP-2), and effectively narrows the performance gap between models trained with either 4M or 129M image-text pairs. Importantly, our framework is modality-agnostic and flexible in terms of architectural design, as validated by its successful application in a video learning task using varied base modules. The code will be made available at https://github.com/yiren-jian/BLIText.

📄 PDF Abstract BibTeX arXiv:2307.07063

Code (1)

yiren-jian/blitext 공식 구현 pytorch

Tasks

Image to text

Methods 이 논문이 사용한 방법론

Focus 설명 없음
BASE 설명 없음
ALIGN In the ALIGN method, visual and language representations are jointly trained from noisy image alt-text data. The image and text encoders are learned via contrastive loss…

Similar Papers 제목 키워드 기반

Mean BERTs make erratic language teachers: the effectiveness of latent bootstrapping in low-resource settings

2023-10-30 · David Samuel

This paper explores the use of latent bootstrapping, an alternative self-supervision technique, for pretraining language models. Unlike the typical practice of using self-supervision on discrete subwords, latent bootstra…

BLIP: Bootstrapping Language-Image Pre-training for Unified Vision-Language Understanding and Generation

2022-01-28 · Junnan Li, Dongxu Li, Caiming Xiong, Steven Hoi

Vision-Language Pre-training (VLP) has advanced the performance for many vision-language tasks. However, most existing pre-trained models only excel in either understanding-based tasks or generation-based tasks. Furtherm…

Image CaptioningImage-text matchingImage-text RetrievalOpen Vocabulary Attribute Detection+4

Dynamic Multimodal Evaluation with Flexible Complexity by Vision-Language Bootstrapping

2024-10-11 · Yue Yang, Shuibai Zhang, Wenqi Shao, Kaipeng Zhang 외

Large Vision-Language Models (LVLMs) have demonstrated remarkable capabilities across multimodal tasks such as visual perception and reasoning, leading to good performance on various multimodal evaluation benchmarks. How…

MMEQuestion AnsweringVisual Question Answering

LDP: Generalizing to Multilingual Visual Information Extraction by Language Decoupled Pretraining

2024-12-19 · Huawen Shen, Gengluo Li, Jinwen Zhong, Yu Zhou

Visual Information Extraction (VIE) plays a crucial role in the comprehension of semi-structured documents, and several pre-trained models have been developed to enhance performance. However, most of these works are mono…

Libra: Building Decoupled Vision System on Large Language Models

2024-05-16 · Yifan Xu, Xiaoshan Yang, Yaguang Song, Changsheng Xu

In this work, we introduce Libra, a prototype model with a decoupled vision system on a large language model (LLM). The decoupled vision system decouples inner-modal modeling and cross-modal interaction, yielding unique …

Image to textLanguage ModelingLanguage ModellingLarge Language Model