[GH-ISSUE #7530] submitting corrections to models. #4789

Closed
opened 2026-04-12 15:45:25 -05:00 by GiteaMirror · 2 comments
Owner

Originally created by @ht990332 on GitHub (Nov 6, 2024).
Original GitHub issue: https://github.com/ollama/ollama/issues/7530

What is the issue?

Hello. I am using LLama3.2 and I asked it an RXJS question. The answer was clear and pleasantly complete but there was a big mistake in the answer. Citation to corrections can be found on RXJS documentation. How do I submit corrections to the model's answers?

OS

Linux

GPU

Intel

CPU

Intel

Ollama version

0.3.14

Originally created by @ht990332 on GitHub (Nov 6, 2024). Original GitHub issue: https://github.com/ollama/ollama/issues/7530 ### What is the issue? Hello. I am using LLama3.2 and I asked it an RXJS question. The answer was clear and pleasantly complete but there was a big mistake in the answer. Citation to corrections can be found on RXJS documentation. How do I submit corrections to the model's answers? ### OS Linux ### GPU Intel ### CPU Intel ### Ollama version 0.3.14
GiteaMirror added the question label 2026-04-12 15:45:25 -05:00
Author
Owner

@rick-github commented on GitHub (Nov 6, 2024):

A model's information is fixed. You can include updates by adding a layer on top (eg, RAG or fine-tuning) that uses fresher data sources, but the mistakes that the model was trained on can't be updated.

<!-- gh-comment-id:2460489932 --> @rick-github commented on GitHub (Nov 6, 2024): A model's information is fixed. You can include updates by adding a layer on top (eg, RAG or fine-tuning) that uses fresher data sources, but the mistakes that the model was trained on can't be updated.
Author
Owner

@pdevine commented on GitHub (Nov 13, 2024):

Unfortunately there's no real feedback mechanism to get this data back to the model creators. You can of course fine tune a model to give it different properties, although we don't have fine tuning yet in Ollama (although you can take a fine tuned adapter that you have created and use it in Ollama).

Thank you for submitting the issue, but I'm going to go ahead and close the issue since there's not much we can do here.

<!-- gh-comment-id:2474907585 --> @pdevine commented on GitHub (Nov 13, 2024): Unfortunately there's no real feedback mechanism to get this data back to the model creators. You can of course fine tune a model to give it different properties, although we don't have fine tuning yet in Ollama (although you can take a fine tuned adapter that you have created and use it in Ollama). Thank you for submitting the issue, but I'm going to go ahead and close the issue since there's not much we can do here.
Sign in to join this conversation.
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: github-starred/ollama#4789