IterIS merges multiple LoRAs into one adapter by iteratively re-estimating input features and solving a closed-form alignment objective, needing only 1-5% of the samples required by prior methods.
Automatically constructing a corpus of sentential paraphrases
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2024 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
IterIS: Iterative Inference-Solving Alignment for LoRA Merging
IterIS merges multiple LoRAs into one adapter by iteratively re-estimating input features and solving a closed-form alignment objective, needing only 1-5% of the samples required by prior methods.