Update README.md

README.md

@@ -11,9 +11,11 @@ licenses:
 
 ## Model overview
 
-Mutual Implication Score: a symmetric measure of text semantic similarity
+Mutual Implication Score is a symmetric measure of text semantic similarity
 based on a RoBERTA model pretrained for natural language inference
-and fine-tuned for paraphrase detection. It is particularly useful for paraphrases detection
+and fine-tuned on paraphrases dataset. 
+
+It is **particularly useful for paraphrases detection**, but can also be applied to other semantic similarity tasks, such as text style transfer.
 
 ## How to use
 The following snippet illustrates code usage:
@@ -33,7 +35,7 @@ print(scores)
 
 ## Model details
 
-We slightly modify [RoBERTa-Large NLI](https://huggingface.co/ynie/roberta-large-snli_mnli_fever_anli_R1_R2_R3-nli) model architectures (see the scheme below) and fin-tune it with [QQP](https://www.kaggle.com/c/quora-question-pairs) paraphrases dataset.
+We slightly modify [RoBERTa-Large NLI](https://huggingface.co/ynie/roberta-large-snli_mnli_fever_anli_R1_R2_R3-nli) model architecture (see the scheme below) and fine-tune it with [QQP](https://www.kaggle.com/c/quora-question-pairs) paraphrases dataset.
 
 ![alt text](https://huggingface.co/SkolkovoInstitute/Mutual_Implication_Score/raw/main/MIS.jpg)
 
@@ -41,7 +43,13 @@ We slightly modify [RoBERTa-Large NLI](https://huggingface.co/ynie/roberta-large
 
 This measure was developed in terms of large scale comparison of different measures on text style transfer and paraphrases datasets.
 
+<img src="https://huggingface.co/SkolkovoInstitute/Mutual_Implication_Score/raw/main/corr_main.jpg" alt="drawing" width="1000"/>
+
+The scheme above shows the correlations of measures of different classes with human judgments on paraphrase and text style transfer datasets. The text above each dataset indicates the best-performing measure. The rightmost columns show the mean performance of measures across the datasets.
+
+MIS outperforms all measures on paraphrases detection task and performs on par with top measures on text style transfer task. 
 
+To learn more refer to our article: [A large-scale computational study of content preservation measures for text style transfer and paraphrase generation](https://aclanthology.org/2022.acl-srw.23/)
 
 ## Citations
 If you find this repository helpful, feel free to cite our publication: