FAAR: Efficient Frequency-Aware Multi-Task Fine-Tuning via Automatic Rank Selection

TL;DR

FAAR introduces a frequency-aware, automatic rank selection method for multi-task fine-tuning that significantly reduces parameters while improving accuracy by learning task-specific spatial and spectral information.

Contribution

The paper proposes FAAR, a novel PEFT approach that adaptively allocates ranks and incorporates spectral analysis for enhanced multi-task learning efficiency and performance.

Findings

Reduces parameters by up to 9 times compared to traditional methods.

Outperforms existing PEFT methods in accuracy on visual benchmarks.

Improves task relationship modeling through spectral pyramid decoding.

Abstract

Adapting models pre-trained on large-scale datasets is a proven way to reach strong performance quickly for down-stream tasks. However, the growth of state-of-the-art mod-els makes traditional full fine-tuning unsuitable and difficult, especially for multi-task learning (MTL) where cost scales with the number of tasks. As a result, recent studies investigate parameter-efficient fine-tuning (PEFT) using low-rank adaptation to significantly reduce the number of trainable parameters. However, these existing methods use a single, fixed rank, which may not be optimal for differ-ent tasks or positions in the MTL architecture. Moreover, these methods fail to learn spatial information that cap-tures inter-task relationships and helps to improve diverse task predictions. This paper introduces Frequency-Aware and Automatic Rank (FAAR) for efficient MTL fine-tuning. Our method introduces Performance-Driven Rank Shrink-ing (PDRS) to allocate the optimal rank per adapter location and per task. Moreover, by analyzing the image frequency spectrum, FAAR proposes a Task-Spectral Pyramidal Decoder (TS-PD) that injects input-specific context into spatial bias learning to better reflect cross-task relationships. Experiments performed on dense visual task benchmarks show the superiority of our method in terms of both accuracy and efficiency compared to other PEFT methods in MTL. FAAR reduces the number of parameters by up to 9 times compared to traditional MTL fine-tuning whilst improving overall performance. Our code is available.