|
--- |
|
license: mit |
|
--- |
|
|
|
Port of the original lilt-only-base model weights from the [Language-Independent Layout Transformer (LiLT)](https://arxiv.org/pdf/2202.13669v1.pdf) |
|
|
|
The weights found here are not useful as a standalone and should be instead used in combination with Roberta-like models as outlined [HERE](https://github.com/jpWang/LiLT#or-generate-your-own-checkpoint-optional) |
|
|
|
This repository aims to make it easier for others to combine LiLT with a Roberta-like model of their liking. |
|
Please refer to the following script on how to fuse XLM-Roberta with LiLT for multi-modal training/fine-tuning [HERE](https://github.com/NielsRogge/Transformers-Tutorials/blob/master/LiLT/Create_LiLT_%2B_XLM_RoBERTa_base.ipynb) |
