Update README.md

README.md

@@ -9,17 +9,17 @@ tags:
 - SMM4H
 
 widget:
-- text: "joints killing me now i have gone back up on the lamotrigine. sick of side effects. sick of meds. want my own self back. knackered today"
-  example_title: "Adverse Drug Event Detection - ADE"
-- text: "bout to have a kick ass summer then it's time to get serious fer school #vyvanse #geekmode"
-  example_title: "Adverse Drug Event Detection - noADE"
+- text: "assert ade: joints killing me now i have gone back up on the lamotrigine. sick of side effects. sick of meds. want my own self back. knackered today"
+  example_title: "ADE"
+- text: "assert ade: bout to have a kick ass summer then it's time to get serious fer school #vyvanse #geekmode"
+  example_title: "noADE"
 ---
 
 ## t2t-assert-ade-balanced
 
 t2t-assert-ade-balanced is a text-to-text (**t2t**) adverse drug event (**ade**) detection model trained with over- and undersampled (balanced) English tweets reporting adverse drug events. It is trained as part of BOUN-TABI system for the Social Media Mining for Health (SMM4H) 2022 shared task. The system description paper has been accepted for publication in *Proceedings of the Seventh Social Media Mining for Health (#SMM4H) Workshop and Shared Task* and will be available soon. The source code has been released on GitHub at [https://github.com/gokceuludogan/boun-tabi-smm4h22](https://github.com/gokceuludogan/boun-tabi-smm4h22).
 
-The model utilizes the T5 model and its text-to-text formulation. The inputs are fed to the model directly as a sentence/tweet, and receives the output "adverse event problem" or "healthy okay". 
+The model utilizes the T5 model and its text-to-text formulation. The inputs are fed to the model with the task prefix "assert ade:", followed with a sentence/tweet. In turn, the output "adverse event problem" or "healthy okay" is received. 
 
 ## Requirements
 ```
@@ -34,7 +34,7 @@ from transformers import pipeline, AutoTokenizer, AutoModelForSeq2SeqLM
 tokenizer = AutoTokenizer.from_pretrained("yirmibesogluz/t2t-assert-ade-balanced")
 model = AutoModelForSeq2SeqLM.from_pretrained("yirmibesogluz/t2t-assert-ade-balanced")
 predictor = pipeline("text2text-generation", model=model, tokenizer=tokenizer)
-predictor("joints killing me now i have gone back up on the lamotrigine. sick of side effects. sick of meds. want my own self back. knackered today")
+predictor("assert ade: joints killing me now i have gone back up on the lamotrigine. sick of side effects. sick of meds. want my own self back. knackered today")
 ```
 
 ## Citation