Update README.md

README.md

@@ -1,23 +1,33 @@
 ---
+language: en
+pipeline_tag: fill-mask
+license: cc-by-sa-4.0
 tags:
-- generated_from_trainer
+- legal
 model-index:
-- name: roberta-large-cased
+- name: lexlms/roberta-large
   results: []
+widget:
+- text: "The applicant submitted that her husband was subjected to treatment amounting to [MASK] whilst in the custody of police."
+datasets:
+- lexlms/lexfiles
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
 should probably proofread and complete it, then remove this comment. -->
 
-# roberta-large-cased
+# LexLM large
 
-This model was trained from scratch on an unknown dataset.
-It achieves the following results on the evaluation set:
-- Loss: 0.6314
+This model was continued pre-trained from RoBERTa large (https://huggingface.co/roberta-large) on the LeXFiles corpus (https://huggingface.co/datasets/lexlms/lexfiles).
 
 ## Model description
 
-More information needed
+LexLM (Base/Large) are our newly released RoBERTa models. We follow a series of best-practices in language model development:
+* We warm-start (initialize) our models from the original RoBERTa checkpoints (base or large) of Liu et al. (2019).
+* We train a new tokenizer of 50k BPEs, but we reuse the original embeddings for all lexically overlapping tokens (Pfeiffer et al., 2021).
+* We continue pre-training our models on the diverse LeXFiles corpus for additional 1M steps with batches of 512 samples, and a 20/30% masking rate (Wettig et al., 2022), for base/large models, respectively. 
+* We use a sentence sampler with exponential smoothing of the sub-corpora sampling rate following Conneau et al. (2019) since there is a disparate proportion of tokens across sub-corpora and we aim to preserve per-corpus capacity (avoid overfitting).
+* We consider mixed cased models, similar to all recently developed large PLMs.
 
 ## Intended uses & limitations
 
@@ -25,7 +35,7 @@ More information needed
 
 ## Training and evaluation data
 
-More information needed
+The model was trained on the LeXFiles corpus (https://huggingface.co/datasets/lexlms/lexfiles). For evaluation results, please consider our work "LeXFiles and LegalLAMA: Facilitating English Multinational Legal Language Model Development" (Chalkidis* et al, 2023).
 
 ## Training procedure
 
@@ -78,3 +88,25 @@ The following hyperparameters were used during training:
 - Pytorch 1.12.0+cu102
 - Datasets 2.7.0
 - Tokenizers 0.12.0
+
+### Citation
+
+[*Ilias Chalkidis\*, Nicolas Garneau\*, Catalina E.C. Goanta, Daniel Martin Katz, and Anders Søgaard.*
+*LeXFiles and LegalLAMA: Facilitating English Multinational Legal Language Model Development.*
+*2022. In the Proceedings of the 61th Annual Meeting of the Association for Computational Linguistics. Toronto, Canada.*](https://aclanthology.org/xxx/)
+```
+@inproceedings{chalkidis-garneau-etal-2023-lexlms,
+    title = {{LeXFiles and LegalLAMA: Facilitating English Multinational Legal Language Model Development}},
+    author = "Chalkidis*, Ilias and 
+              Garneau*, Nicolas and
+              Goanta, Catalina and 
+              Katz, Daniel Martin and 
+              Søgaard, Anders",
+    booktitle = "Proceedings of the 61h Annual Meeting of the Association for Computational Linguistics",
+    month = june,
+    year = "2023",
+    address = "Toronto, Canada",
+    publisher = "Association for Computational Linguistics",
+    url = "https://aclanthology.org/xxx",
+}
+```