Originally created by @Snify89 on GitHub (Oct 3, 2024).
Is your feature request related to a problem? Please describe.
No. This is just an idea I'd like to have implemented.
Describe the solution you'd like
It would be great to have an option to "hide" the references coming from the RAG. If you use doc search, you see the references and the clear text from the document. There might be cases, where this is unwanted. E.g. I have "sensitive" contents that the LLM should filter out, yet I can see the entire documents.
Describe alternatives you've considered
I have a workaround by feeding the system prompt with the most important content needed. This needs a higher context size.
Additional context
I hope you understand what I mean.
Thank you for reading and considering my idea/suggestion. Keep up your good work. :)
Originally created by @Snify89 on GitHub (Oct 3, 2024).
**Is your feature request related to a problem? Please describe.**
No. This is just an idea I'd like to have implemented.
**Describe the solution you'd like**
It would be great to have an option to "hide" the references coming from the RAG. If you use doc search, you see the references and the clear text from the document. There might be cases, where this is unwanted. E.g. I have "sensitive" contents that the LLM should filter out, yet I can see the entire documents.
**Describe alternatives you've considered**
I have a workaround by feeding the system prompt with the most important content needed. This needs a higher context size.
**Additional context**
I hope you understand what I mean.
Thank you for reading and considering my idea/suggestion. Keep up your good work. :)
Not sure this is really what you want, if the AI has access to the content, there's always someone who will find a way to make the AI rewrite that content verbatim. This works until it doesn't, removing the sensitive information might be the only solution there.
@ticpu commented on GitHub (Oct 4, 2024):
Not sure this is really what you want, if the AI has access to the content, there's always someone who will find a way to make the AI rewrite that content verbatim. This works until it doesn't, removing the sensitive information might be the only solution there.
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.
Originally created by @Snify89 on GitHub (Oct 3, 2024).
Is your feature request related to a problem? Please describe.
No. This is just an idea I'd like to have implemented.
Describe the solution you'd like
It would be great to have an option to "hide" the references coming from the RAG. If you use doc search, you see the references and the clear text from the document. There might be cases, where this is unwanted. E.g. I have "sensitive" contents that the LLM should filter out, yet I can see the entire documents.
Describe alternatives you've considered
I have a workaround by feeding the system prompt with the most important content needed. This needs a higher context size.
Additional context
I hope you understand what I mean.
Thank you for reading and considering my idea/suggestion. Keep up your good work. :)
@ticpu commented on GitHub (Oct 4, 2024):
Not sure this is really what you want, if the AI has access to the content, there's always someone who will find a way to make the AI rewrite that content verbatim. This works until it doesn't, removing the sensitive information might be the only solution there.