Update README.md

README.md

@@ -46,9 +46,7 @@ We use the pretrained [`sentence-transformers/all-mpnet-base-v2`](https://huggin
  However, we reduce the processed sequence length to 384 to reduce the training time of the model. 
  Regarding other hyperparameters, we use the same interaction data batch size of 1024; we use the negative sampling parameter m = 10000.
  We use constant learning rate of 1e-5, and we train the model for ten epochs.
- For deatils please see our paper (link TBA).
-
- We finetuned our model on both Goodbooks-10k and MovieLens-20M datasets. 
+ We finetuned our model on both Goodbooks-10k and MovieLens-20M datasets. For deatils please see our paper (link TBA). 
  For item ids used during training please see (links TBA).
 
 ## Evaluation Results