Add paper link to README

README.md

@@ -14,6 +14,23 @@ metrics:
   - rmse
 ---
 
+# Example usage
+```
+from transformers import BertForMaskedLM, PreTrainedTokenizerFast
+
+# Load the tokenizer and model
+tokenizer = PreTrainedTokenizerFast.from_pretrained('thaonguyen217/farm_molecular_representation')
+model = BertForMaskedLM.from_pretrained('thaonguyen217/farm_molecular_representation')
+
+# Example usage
+input_text = "N_primary_amine N_secondary_amine c_6-6 1 n_6-6 n_6-6 c_6-6 c_6-6 2 c_6-6 c_6-6 c_6-6 c_6-6 c_6-6 1 2" # FG-enhanced representation of NNc1nncc2ccccc12
+inputs = tokenizer(input_text, return_tensors='pt')
+outputs = model(**inputs, output_hidden_states=True)
+
+# Extract atom embeddings from last hidden states
+last_hidden_states = outputs.hidden_states[-1][0] # last_hidden_states: (N, 768) with N is input length
+```
+
 # Farm Molecular Representation Model
 You can read more about the model in our [paper](https://arxiv.org/pdf/2410.02082) or [webpage](https://thaonguyen217.github.io/farm/).