I switched to ARM (Windows on ARM) and Socrates runs through the system emulator, but it doesn't look good. There is no file preview (only through the sources in the answers). Your application support is amazing! The ARM version is urgently needed for me - I know it's on the roadmap, but when can we realistically expect it?
Hi, Does it matter to Socrates what filetype the documents are in?
Like pdf versus Word versus excel or epub or Markdown?
And what limits are there? I mean what if I have textbook pdf with like thousand pages. In other ai tool I splitted that document into like 12 chapters to prevent ai hallucinating. Is that something you should do with Socrates too? And if so, at what size should you do that?
A: Hello, Socrates supports all those document types, but PDFs have a bit more structure and thus can provide additional context to LLMs to give a slightly better result (we auto-convert DOCX files to PDF during Deep Dives).
Yes, we handle auto-splitting of long PDFs, so you don't have to worry about the size of the documents.
Q: Inquiry About Extracting Product Descriptions from PDFs Using Socrates
I would like to know if it's possible to extract all sentences from a PDF containing product descriptions using Socrates. The extracted sentences should remain unaltered and include reference numbers (if available).
I attempted to do this and initially received results that I was satisfied with. However, I later discovered that some sentences contained hallucinated information, including fabricated sentences and references that were not present in the original PDF.
I used the "Deep Dive Tools" and "Prompt Loop - Section" features for this task, but I couldn’t find specific instructions regarding this issue. Do you have any recommendations or guidelines on how to handle this problem effectively?
A: Hello -- I saw your support email on this topic so will answer here.
Yes, the best way to do this is: 1) run a Prompt Loop separated by paragraph 2) have the prompt be something like: "Extract all product descriptions in this paragraph. If there are no product descriptions, say 'No Product Description'".
A: Hello! 1. Chinese is already supported (both simplified and traditional): https://roadmap.asksocrates.app/p/set-language-preference-oF6HoY 2. Thanks, please vote on this in the wishlist: https://roadmap.asksocrates.app/t/wish-list 3. What do you mean Humanization?
Q: ARM version
I switched to ARM (Windows on ARM) and Socrates runs through the system emulator, but it doesn't look good. There is no file preview (only through the sources in the answers). Your application support is amazing! The ARM version is urgently needed for me - I know it's on the roadmap, but when can we realistically expect it?
Jon_Socrates
Nov 23, 2024A: We've just bought a Windows ARM computer to test our builds... so please wait a few weeks!
Share Socrates
Q: Prefered format source documents
Hi,
Does it matter to Socrates what filetype the documents are in?
Like pdf versus Word versus excel or epub or Markdown?
And what limits are there? I mean what if I have textbook pdf with like thousand pages. In other ai tool I splitted that document into like 12 chapters to prevent ai hallucinating. Is that something you should do with Socrates too? And if so, at what size should you do that?
Jon_Socrates
Nov 23, 2024A: Hello, Socrates supports all those document types, but PDFs have a bit more structure and thus can provide additional context to LLMs to give a slightly better result (we auto-convert DOCX files to PDF during Deep Dives).
Yes, we handle auto-splitting of long PDFs, so you don't have to worry about the size of the documents.
Share Socrates
Q: Inquiry About Extracting Product Descriptions from PDFs Using Socrates
I would like to know if it's possible to extract all sentences from a PDF containing product descriptions using Socrates. The extracted sentences should remain unaltered and include reference numbers (if available).
I attempted to do this and initially received results that I was satisfied with. However, I later discovered that some sentences contained hallucinated information, including fabricated sentences and references that were not present in the original PDF.
I used the "Deep Dive Tools" and "Prompt Loop - Section" features for this task, but I couldn’t find specific instructions regarding this issue. Do you have any recommendations or guidelines on how to handle this problem effectively?
Jon_Socrates
Nov 23, 2024A: Hello --
I saw your support email on this topic so will answer here.
Yes, the best way to do this is:
1) run a Prompt Loop separated by paragraph
2) have the prompt be something like: "Extract all product descriptions in this paragraph. If there are no product descriptions, say 'No Product Description'".
^this should minimize hallucinations.
Share Socrates
Q: add the function
1. Add the language "Chinese".
2. Add the function of writing long articles in various styles.
3，Humanization
thank you
Jon_Socrates
Nov 22, 2024A: Hello!
1. Chinese is already supported (both simplified and traditional): https://roadmap.asksocrates.app/p/set-language-preference-oF6HoY
2. Thanks, please vote on this in the wishlist: https://roadmap.asksocrates.app/t/wish-list
3. What do you mean Humanization?
Share Socrates
Q: will you be adding voice input ? and or speech output ?
tx
Jon_Socrates
Nov 22, 2024A: Hello! I see this has been added to the wish list. Will ship depending on community votes!
Share Socrates