I’ve been reading about sentence similarity on the huggingface website (namely here). I have a very basic question: What is the difference between Passage Ranking and Semantic Textual Similarity?
To my untrained eye, it appears that both tasks are doing the same thing but utilizing different ranking metrics, i.e. calculate the similarity between a query and your document database and rank the resulting list based on those metrics. However, seeing as they are two separate tasks, my assumption is that they tasks are looking at unique aspects of Sentence Similarity.