Add Arrow + GenKnowSub to LoRA #2644
Open
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Hi there! 👋
This PR adds support for Arrow, a modular routing mechanism for LoRA experts introduced here, as well as our refinement method GenKnowSub, proposed in our ACL 2025 Main Conference paper. GenKnowSub enhances Arrow by subtracting a general-domain LoRA from task-specific ones prior to routing, leading to improved generalisation and modularity.
The integration is implemented through a new ArrowLoraLinearLayer, registered as a LoraVariant (similar to DoRA).
🔧 Code Changes
Modified files under peft/tuners/lora/:
Added a new file:
✅ Testing & Quality
Confirmed that tests/test_lora_variants.py passes.
All formatting and style checks pass (make quality successful).
📚 Optional Follow-up
I’ve also developed a separate GitHub repository demonstrating how to use Arrow and GenKnowSub on benchmark datasets (e.g., BoolQ, PIQA, ARC-Challenge). I’d be happy to contribute this as an example under peft/examples/ if it’s of interest.
Thank you in advance for reviewing this PR!