readme

README.md

@@ -13,7 +13,8 @@ pipeline_tag: fill-mask
 **This model relies on a custom modeling file, you need to add trust_remote_code=True**\
 **See [\#13467](https://github.com/huggingface/transformers/pull/13467)**
 
-Conversion script is available at this [link](https://github.com/ccdv-ai/convert_checkpoint_to_lsg).
+LSG ArXiv [paper](https://arxiv.org/abs/2210.15497). \
+Github/conversion script is available at this [link](https://github.com/ccdv-ai/convert_checkpoint_to_lsg).
 
 * [Usage](#usage)
 * [Parameters](#parameters)
@@ -25,7 +26,7 @@ This model is adapted from [BART-base](https://huggingface.co/facebook/bart-base
 
 This model can handle long sequences but faster and more efficiently than Longformer (LED) or BigBird (Pegasus) from the hub and relies on Local + Sparse + Global attention (LSG).
 
-The model requires sequences whose length is a multiple of the block size. The model is "adaptive" and automatically pads the sequences if needed (adaptive=True in config). It is however recommended, thanks to the tokenizer, to truncate the inputs (truncation=True) and optionally to pad with a multiple of the block size (pad_to_multiple_of=...). \
+The model requires sequences whose length is a multiple of the block size. The model is "adaptive" and automatically pads the sequences if needed (adaptive=True in config). It is however recommended, thanks to the tokenizer, to truncate the inputs (truncation=True) and optionally to pad with a multiple of the block size (pad_to_multiple_of=...). 
 
 Implemented in PyTorch.