LoRA Recycle distills a meta-LoRA from diverse pre-tuned LoRAs using data generated by inverting those LoRAs, enabling tuning-free few-shot classification at test time.
Meta-adapters: Parameter ef- ficient few-shot fine-tuning through meta-learning
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.CV 1years
2024 1verdicts
CONDITIONAL 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
Unlocking Tuning-Free Few-Shot Adaptability in Visual Foundation Models by Recycling Pre-Tuned LoRAs
LoRA Recycle distills a meta-LoRA from diverse pre-tuned LoRAs using data generated by inverting those LoRAs, enabling tuning-free few-shot classification at test time.