1–10 of 53 posts
Show HN: ChatGPT and Document Parser = Ghost
ghostextension.com
In my testing the biggest challenges with using for example OpenAI embeddings with cosine similarity or something are A) figuring out the section breaks or right chunk size so that information stays in context and B) retrieving enough chunks to get the correct hit for a query without having too much extraneous information that confuses it.
I think that it's hard to make a parser that most optimally slices up arbitrary documents.
Since you have some larger documents preloaded I assume for those you have the embeddings search. But for user uploads you are skipping that now and just feeding all of the text extracted from the PDF into the prompt along with the query.
I didn't do any extensive testing but seems to be really useful. However, where can we see the privacy policy? People are probably going to upload some important and confidential documents so it's good to know how this data is being handled. The only thing I see is an asterisk after the 24 hour notice Also, the bot answer window may have 5 pages even if 1 page is enough for the answer, this may confise your users sin…