SAN: Hypothesizing Long-Term Synaptic Development and Neural Engram Mechanism in Scalable Model's Parameter-Efficient Fine-Tuning

TL;DR

This paper introduces SAN, a biologically inspired method for parameter-efficient fine-tuning that decomposes and propagates scaling components, significantly improving performance across vision, language, and multimodal tasks.

Contribution

SAN is a novel fine-tuning approach grounded in neurobiological principles, connecting low-rank parameter space shifts with synaptic development mechanisms, and demonstrating superior results over existing methods.

Findings

Outperforms FFT by up to 8.7% on vision tasks

Surpasses ChatGPT by 8.5% on language tasks

Exceeds LoRA by 3.2% on vision and 4.7% on language tasks

Abstract

Advances in Parameter-Efficient Fine-Tuning (PEFT) bridged the performance gap with Full Fine-Tuning (FFT) through sophisticated analysis of pre-trained parameter spaces. Starting from drawing insights from Neural Engrams (NE) in Biological Neural Networks (BNNs), we establish a connection between the low-rank property observed during PEFT's parameter space shifting and neurobiological mechanisms. This observation leads to our proposed method, Synapse and Neuron (SAN), which decomposes and propagates scaling components from anterior feature adjusting vectors towards posterior weight matrices. Our approach is theoretically grounded in Long-Term Potentiation/Depression (LTP/D) phenomena, which govern synapse development through neurotransmitter release modulation. Extensive experiments demonstrate its effectiveness: on \textbf{vision tasks} across VTAB, FGVC, and GIC (25 datasets) using ViT, SwinT and ConvNeXt, SAN outperforms FFT up to 8.7% and LoRA by 3.2%; on language tasks using Commonsense Reasoning (8 datasets) with LLaMA models (all generations), surpassing ChatGPT up to 8.5% and LoRA by 4.7%; on visual-language tasks using Mixed Visual Instruction (7 datasets) with LLaVA models, it exceeds FFT up to 2.4% and LoRA by 1.9%. Our code and W&B log will be released.