1,151
edits
AskYourPDF (ChatGPT Plugin): Difference between revisions
no edit summary
No edit summary |
Startledcat (talk | contribs) No edit summary |
||
| Line 28: | Line 28: | ||
#Please highlight the parts of this PDF: ${pdf_url} that are relevant to ${topic}. | #Please highlight the parts of this PDF: ${pdf_url} that are relevant to ${topic}. | ||
#In which section of this document: ${pdf_url} is ${topic} discussed? | #In which section of this document: ${pdf_url} is ${topic} discussed? | ||
===Example Usage=== | |||
====Prompt==== | |||
<pre> | |||
</pre> | |||
====ChatGPT Response==== | |||
<pre> | |||
The PDF you provided is a paper that proposes a new network architecture called the Transformer. This architecture is based solely on attention mechanisms and does not use recurrent or convolutional neural networks. The Transformer is shown to be superior in quality, more parallelizable, and requires significantly less time to train compared to existing models. The model achieves state-of-the-art results on machine translation tasks and generalizes well to other tasks such as English constituency parsing. The authors attribute the success of the Transformer to the use of attention mechanisms. | |||
If you have more specific questions about the content of the paper, feel free to ask! | |||
Also, please note that you can delete your document at any point by visiting this link. | |||
</pre> | |||
====Plugin Request and Response==== | |||
'''REQUEST TO SCHOLAR ASSIST''' | |||
<pre> | |||
{ | |||
"url": "https://arxiv.org/pdf/1706.03762.pdf" | |||
} | |||
</pre> | |||
'''RESPONSE FROM SCHOLAR ASSIST''' | |||
<pre> | |||
{ | |||
"docId": "1f701426-ec4c-43a2-86da-416c57d2a3d3", | |||
"summary": "The paper proposes a new network architecture called the Transformer, which is based solely on attention mechanisms and does not use recurrent or convolutional neural networks. The Transformer is shown to be superior in quality, more parallelizable, and requires significantly less time to train compared to existing models. The model achieves state-of-the-art results on machine translation tasks and generalizes well to other tasks such as English constituency parsing. The authors attribute the success of the Transformer to the use of attention mechanisms.", | |||
"extra_instructions": "Inform the user they can delete their document at any point by visiting https://askyourpdf.com/delete/{APPEND_DOC_ID_HERE}. Also prompt the user to ask more questions" | |||
} | |||
</pre> | |||
==Tips and Tricks== | ==Tips and Tricks== | ||