baidu/MonoGPT

This repository contains the official checkpoint for PixelGPT, as presented in the paper Autoregressive Pre-Training on Pixels and Texts (EMNLP 2024). For detailed instructions on how to use the model, please visit our GitHub page.

Model Description

MonoGPT is an autoregressive language model pre-trained on the dual modality of both pixels and texts without relying on the pixel-text paired data. By processing documents as visual data (pixels), the model learns to predict both the next token and the next image patch in a sequence, enabling it to handle visually complex tasks in different modalities.

Citation

@misc{chai2024autoregressivepretrainingpixelstexts,
  title = {Autoregressive Pre-Training on Pixels and Texts},
  author = {Chai, Yekun and Liu, Qingyi and Xiao, Jingwu and Wang, Shuohuan and Sun, Yu and Wu, Hua},
  year = {2024},
  eprint = {2404.10710},
  archiveprefix = {arXiv},
  primaryclass = {cs.CL},
  url = {https://arxiv.org/abs/2404.10710},
}