GLT-PEFT: Gated Lie-Tucker Parameter-Efficient Fine-Tuning for Alzheimer's Disease Diagnosis with Hippocampal Segmentation Pretraining

2026-05-15

AI Summary

This work addresses the limitations of existing parameter-efficient fine-tuning methods in adapting to high-dimensional 3D convolutional kernels prevalent in medical imaging and their inability to preserve the geometric structure of pretrained parameters. To overcome these challenges, we propose a geometry-aware efficient fine-tuning framework that, for the first time, unifies additive and multiplicative updates within a single architecture. Specifically, we employ Tucker tensor decomposition to enable low-rank adaptation of 3D convolutional kernels, integrate Lie group transformations to maintain the manifold structure of pretrained parameters, and introduce a gating mechanism to fuse multi-path updates. The resulting approach drastically reduces the number of trainable parameters while demonstrating superior performance and robustness when transferring a hippocampus segmentation pretrained model to Alzheimer’s disease classification tasks.

Abstract

Parameter-efficient fine-tuning (PEFT) has emerged as a promising paradigm for adapting pretrained models under limited data conditions. However, most existing PEFT methods are designed for matrix-structured parameters and are not well suited for high-dimensional convolutional kernels in medical imaging models. Moreover, they typically rely on additive updates and lack mechanisms to preserve the geometric structure of pretrained parameters, while multiplicative (geometry-aware) updates are difficult to integrate within a unified framework. To address this issue, this paper proposes GLT-PEFT, a gated Lie-Tucker parameter-efficient fine-tuning framework for Alzheimer's disease (AD) diagnosis. The proposed approach transfers a hippocampal segmentation pretrained model to a downstream classification task. Tucker decomposition enables tensor-aware low-rank adaptation of 3D convolutional kernels, while Lie group-based transformations provide structure-preserving multiplicative updates. A gating mechanism further reconciles additive and multiplicative update forms, resulting in a unified and more stable fine-tuning strategy. Extensive experiments demonstrate that GLT-PEFT achieves effective cross-task transfer while significantly reducing trainable parameters, highlighting its effectiveness for efficient and robust adaptation in medical imaging models.

Problem

• Parameter-Efficient Fine-Tuning
• Medical Imaging
• High-Dimensional Convolutional Kernels
• Geometric Structure Preservation
• Alzheimer's Disease Diagnosis

Innovation

• Parameter-Efficient Fine-Tuning
• Tucker Decomposition
• Lie Group
• Medical Imaging
• 3D Convolutional Kernels