Update app.py

app.py

@@ -22,12 +22,10 @@ def test_input(words):
     return word_dict
 
     
-title = "BERT on a PLANE"
+title = "Phrase-Entailment Detection with BERT"
 
 description = """
-Did you know that logically speaking **A small cat is not a small animal**, and that **A fake smile is not a smile**? Learn more by testing our BERT model tuned to perform phrase-level adjective-noun entailment, via the [PLANE](https://aclanthology.org/2022.coling-1.359/) dataset.
-
-Please note that the scope of the model is not to run lexical entailment or hypernym detection (e.g., *"A dog is an animal*"), but to perform a very specific subset of phrase-level entailment, based on adjective-nouns phrases. The type of question you can ask the model are limited, and should have one of three forms:
+Did you know that logically speaking **A small cat is not a small animal**, and that **A fake smile is not a smile**? Learn more by testing our BERT model tuned to perform phrase-level adjective-noun entailment. The proposed model was tuned with a section of the PLANE (**P**hrase-**L**evel **A**djective-**N**oun **E**ntailment) dataset, introduced in COLING 2022 [Bertolini et al.,](https://aclanthology.org/2022.coling-1.359/). Please note that the scope of the model is not to run lexical-entailment or hypernym detection (e.g., *"A dog is an animal*"), but to perform a very specific subset of phrase-level compositional entailment over adjective-noun phrases. The type of question you can ask the model are limited, and should have one of three forms:
 
 - An *Adjective-Noun* is a *Noun* (e.g. A red car is a car)
 
@@ -35,7 +33,7 @@ Please note that the scope of the model is not to run lexical entailment or hype
 
 - An *Adjective-Noun* is a *Adjective-Hypernym(Noun)* (e.g. A red car is a red vehicle)
 
-Please note that, as in the examples above, the adjective should be the same for both phrases, and that the Hypernym(Noun) should be a true hypernym of the selected noun.
+As in the examples above, the **adjective should be the same for both phrases**, and the **Hypernym(Noun) should be a true hypernym of the selected noun**.
 
 The current model achieves an accuracy of 90% on out-of-distribution evaluation.
 Use the next page to check if your test-items (i.e. adjective, noun and hypernyms) were part of the training data!"""