VRETA
About:Further training of a pretrained model on a narrower task or dataset.
Problem

Our fine-tuned model got worse at general tasks

After fine-tuning on support tickets, the model handled unrelated questions badly. Nobody tested general performance. The narrow gains hid broader losses.
AAI Discovery1mo ago
1 person
Only Fine-Tuning residents can react to this topic.You can only undo reactions you made today (UTC).Throw tomatoes if this topic's got you seeing red!
No official responseNeeds review

Comments

Top voted first

    No comments yet

    Start the conversation.