Less is More: Resource-Efficient Low-Rank Adaptation
By: Chunlin Tian , Xuyang Wei , Huanrong Liu and more
Potential Business Impact:
Makes AI learn faster and better with less effort.
Low-Rank Adaptation (LoRA) is a widely adopted parameter-efficient fine-tuning (PEFT) method for Large Language Models (LLMs), but it still incurs notable overhead and suffers from parameter interference in complex datasets. While re- cent works decouple LoRA update matrices to exploit matrix-wise asymmetry, training costs remain high. We revisit LoRA from the perspective of inter-matrix and intra-layer parameter redundancy and propose Resource-Efficient Low-Rank Adaptation, EffiLoRA, a lightweight and generalizable approach for language, multimodal, and diffusion models. EffiLoRA employs a unified A matrix across all transformer layers and introduces a runtime selective B matrices up- date to dynamically trade-off the system resource budget and model performance. EffiLoRA consistently outperforms LoRA across diverse modalities, including commonsense reasoning, visual instruction tuning, and image generation, demon- strating improved efficiency and robustness.
Similar Papers
Sensitivity-LoRA: Low-Load Sensitivity-Based Fine-Tuning for Large Language Models
Machine Learning (CS)
Makes smart computer programs learn faster and better.
DropLoRA: Sparse Low-Rank Adaptation for Parameter-Efficient Fine-Tuning
Computation and Language
Makes AI smarter without more training.
ScaLoRA: Optimally Scaled Low-Rank Adaptation for Efficient High-Rank Fine-Tuning
Machine Learning (CS)
Makes smart computer programs learn faster and better.