Update README.md

README.md

@@ -5,6 +5,15 @@ language:
 ---
 This is the proposition segmentation model from "Sub-Sentence Encoder: Contrastive Learning of Propositional Semantic Representations" by Chen et. al. 2023. 
 
+## What does the model do?
+It splits a complex, long-form sentence into a list of propositions -- i.e. self-contained, atomic pieces of meaning in the sentence. For example, the following sentence --
+```
+"Dracula is a novel by Bram Stoker featuring Count Dracula as the protagonist."
+```
+will be split into --
+```
+['Dracula is a novel by Bram Stoker.', 'Count Dracula is the protagonist of Dracula.']
+```
 ## Usage
 The prompt to the model is formatted like: `segment sentence: {input_sentence}`.