[GH-ISSUE #595] [FEAT]: RHLF like feature. #346

Closed
opened 2026-02-22 18:19:04 -05:00 by yindo · 4 comments
Owner

Originally created by @sumitsodhi88 on GitHub (Jan 13, 2024).
Original GitHub issue: https://github.com/Mintplex-Labs/anything-llm/issues/595

Originally assigned to: @shatfield4 on GitHub.

What would you like to see?

An option for Vetting the queries-answers by different users and send it to the workspace embedding. I have this immense resource of question answers that can be used further to train the LLM - although I don't want to go fine-tuning way.

An option for the user to give thumbs-up or down; visible to the manager/ administrator in Workspace Chat pane - and the Administrator can send (after some editing) this to workspace embeddings so that some kind of RHLF happens. But these should be kept separate from the documents uploaded - for better folder management. Thanks

Originally created by @sumitsodhi88 on GitHub (Jan 13, 2024). Original GitHub issue: https://github.com/Mintplex-Labs/anything-llm/issues/595 Originally assigned to: @shatfield4 on GitHub. ### What would you like to see? An option for Vetting the queries-answers by different users and send it to the workspace embedding. I have this immense resource of question answers that can be used further to train the LLM - although I don't want to go fine-tuning way. An option for the user to give thumbs-up or down; visible to the manager/ administrator in Workspace Chat pane - and the Administrator can send (after some editing) this to workspace embeddings so that some kind of RHLF happens. But these should be kept separate from the documents uploaded - for better folder management. Thanks
yindo added the enhancementfeature request labels 2026-02-22 18:19:04 -05:00
yindo closed this issue 2026-02-22 18:19:04 -05:00
Author
Owner

@sumitsodhi88 commented on GitHub (Jan 14, 2024):

Or what could be the other best thing is that let the workspace learn from all the users' chats - this will continuously improve the answers and becomes highly effiecient.

@sumitsodhi88 commented on GitHub (Jan 14, 2024): Or what could be the other best thing is that let the workspace learn from all the users' chats - this will continuously improve the answers and becomes highly effiecient.
Author
Owner

@sumitsodhi88 commented on GitHub (Jan 18, 2024):

But since, the json are not supported to be uploaded this resource is not being utilized well by the application.

@sumitsodhi88 commented on GitHub (Jan 18, 2024): But since, the json are not supported to be uploaded this resource is not being utilized well by the application.
Author
Owner
@sumitsodhi88 commented on GitHub (Jan 28, 2024): https://www.marktechpost.com/2024/01/22/this-ai-paper-from-meta-and-nyu-introduces-self-rewarding-language-models-that-are-capable-of-self-alignment-via-judging-and-training-on-their-own-generations/?amp
Author
Owner

@sumitsodhi88 commented on GitHub (Feb 13, 2024):

brilliant

@sumitsodhi88 commented on GitHub (Feb 13, 2024): brilliant
yindo changed title from [FEAT]: RHLF like feature. to [GH-ISSUE #595] [FEAT]: RHLF like feature. 2026-06-05 14:34:50 -04:00
Sign in to join this conversation.
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: Mintplex-Labs/anything-llm#346