JoLA jointly learns sparse attention-head selection and additive/multiplicative activation edits, outperforming LoRA and prior activation-editing baselines in low-resource LLM fine-tuning.
Examining modularity in multilingual LM s via language-specialized subnetworks
1 Pith paper cite this work, alongside 1 external citations. Polarity classification is still indexing.
1
Pith paper citing it
1
external citations · OpenAlex
fields
cs.CL 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Joint Localization and Activation Editing for Low-Resource Fine-Tuning
JoLA jointly learns sparse attention-head selection and additive/multiplicative activation edits, outperforming LoRA and prior activation-editing baselines in low-resource LLM fine-tuning.