Assistant Uploads

Hey there.

I'm, Drew Barontini, VP of Product for Scite, and I want to talk about a new feature we have called Assistant Uploads.

Now, in the Assistant experience, you're going to see a new plus menu on the prompt, and when you click that, you're going to have a couple of options, the first of which is to attach a collection or multiple collections to the chat, which is a new feature that allows you to collect research and papers within a specific topic, or optionally, which I'm going to show here.

Upload a document.

Now, before, when we uploaded documents to Assistant, it had to go through our reference check feature, which runs a full processing, to look at all the citations and smart citations around anything mentioned inside of that paper.

Now, you can upload any PDF or docx file and have a conversation with it.

So, so I'm going to go ahead and upload a file here.

I'm going to drag and drop this one in here, and then you're going to see that this document is going to be attached and be uploaded to the Assistant session.

Okay, now that we have our document attached here, I'm going to ask a question to say, summarize this.

We'll say the key findings from this paper and then we're going to hit enter here.

And you can see that down here at the bottom we have this attached PDF file attached to this Assistant conversation.

You can see that it's finalizing an answer based on that one source instead of looking at multiple publications across our data set.

So this is going to show this individual findings here.

Now, the PDF I uploaded is actually the paper that references how cite and the smart citations work as a published paper that we have.

So this is going to go through and give a response based on that information.

Okay.

so we have our response here and you can see the inline citations.

Now we're all referencing this PDF file.

So it's actually referencing the PDF and it's pulling out the information because this is the source of that, these are the authors, this is the title, these are the snippets of information that are being pulled from.

This is why it's being selected, because all the claims are tied to that specific paper.

So you can ask questions about this and go in deeper if you need to.

And if we open up the references here, you're going to see the same references here across that PDF file with any information.

So since this is a paper, it's going to show additional data, any other citations that are related to this that are pulling across the rich network that we have access to, showing what section it comes from.

And then there's not much in the search strategy because again, it was just referencing that single file.

So this is a way that you can upload PDFs or documents to the assistant experience to have conversations.

And this is something that can provide more rich outputs for you as you're working, working through your research.

We hope you enjoy it.

And thanks.