Density Adaptive Attention is All You Need: Robust Parameter-Efficient Fine-Tuning Across Multiple Modalities

TL;DR

This paper introduces a probabilistic attention mechanism called DAAM that improves parameter-efficient fine-tuning across speech, text, and vision modalities, achieving significant accuracy gains and better adaptability for non-stationary data.

Contribution

It presents the novel DAAM framework and the Density Adaptive Transformer, advancing multi-modal attention with probabilistic modeling and improved explainability.

Findings

Up to 20% accuracy improvement over state-of-the-art methods

Superior performance on speech, image, and text classification tasks

Enhanced robustness and interpretability with the Importance Factor

Abstract

We propose the Multi-Head Density Adaptive Attention Mechanism (DAAM), a novel probabilistic attention framework that can be used for Parameter-Efficient Fine-tuning (PEFT), and the Density Adaptive Transformer (DAT), designed to enhance information aggregation across multiple modalities, including Speech, Text, and Vision. DAAM integrates learnable mean and variance into its attention mechanism, implemented in a multi-head framework, enabling it to collectively model any probability distribution for dynamic recalibration of feature significance. This method demonstrates significant improvements, especially with highly non-stationary data, surpassing the state-of-the-art attention techniques in model performance, up to approximately +20% (abs.) in accuracy. Empirically, DAAM exhibits superior adaptability and efficacy across a diverse range of tasks, including emotion recognition in speech, image classification, and text classification, thereby establishing its robustness and versatility in handling data across multiple modalities. Furthermore, we introduce the Importance Factor, a new learning-based metric that enhances the explainability of models trained with DAAM-based methods.