Update README.md

README.md

@@ -57,6 +57,11 @@ It achieves the following results on the evaluation set:
 - F1: 0.8584
 - Accuracy: 0.8584
 
+**References:**
+- Blog Post: [Document AI | Document Understanding model at line level with LiLT, Tesseract and DocLayNet dataset]()
+- Notebook: [Document AI | Fine-tune LiLT on DocLayNet base in any language at line level (chunk of 384 tokens with overlap)](https://github.com/piegu/language-models/blob/master/Fine_tune_LiLT_on_DocLayNet_base_in_any_language_at_linelevel_ml_384.ipynb)
+- Notebook: [Document AI | Inference at line level with a Document Understanding model (LiLT fine-tuned on DocLayNet dataset)](https://github.com/piegu/language-models/blob/master/inference_on_LiLT_model_finetuned_on_DocLayNet_base_in_any_language_at_levellines_ml384.ipynb)
+
 ### DocLayNet dataset
 
 [DocLayNet dataset](https://github.com/DS4SD/DocLayNet) (IBM) provides page-by-page layout segmentation ground-truth using bounding-boxes for 11 distinct class labels on 80863 unique pages from 6 document categories. 
@@ -75,11 +80,11 @@ At inference time, a calculation of best probabilities give the label to each li
 
 ## Inference
 
-See notebook: [inference_on_LiLT_model_finetuned_on_DocLayNet_base_in_any_language_at_levellines_ml384.ipynb]()
+See notebook: [Document AI | Inference at line level with a Document Understanding model (LiLT fine-tuned on DocLayNet dataset)](https://github.com/piegu/language-models/blob/master/inference_on_LiLT_model_finetuned_on_DocLayNet_base_in_any_language_at_levellines_ml384.ipynb)
 
 ## Training and evaluation data
 
-See notebook: [Fine_tune_LiLT_on_DocLayNet_base_in_any_language_at_linelevel_ml_384.ipynb]()
+See notebook: [Document AI | Fine-tune LiLT on DocLayNet base in any language at line level (chunk of 384 tokens with overlap)](https://github.com/piegu/language-models/blob/master/Fine_tune_LiLT_on_DocLayNet_base_in_any_language_at_linelevel_ml_384.ipynb)
 
 ## Training procedure