Learning Sparsity for Effective and Efficient Music Performance Question Answering
This work addresses inefficiencies in multimodal reasoning for music performance analysis, offering incremental improvements in efficiency and data usage for domain-specific applications.
The paper tackled the challenge of inefficient multimodal representation in Music Performance Audio-Visual Question Answering by introducing Sparsify, a sparse learning framework that achieved state-of-the-art performance, reduced training time by 28.32%, and maintained accuracy with a key-subset selection algorithm using 25% of training data while retaining 70-80% performance.
Music performances, characterized by dense and continuous audio as well as seamless audio-visual integration, present unique challenges for multimodal scene understanding and reasoning. Recent Music Performance Audio-Visual Question Answering (Music AVQA) datasets have been proposed to reflect these challenges, highlighting the continued need for more effective integration of audio-visual representations in complex question answering. However, existing Music AVQA methods often rely on dense and unoptimized representations, leading to inefficiencies in the isolation of key information, the reduction of redundancy, and the prioritization of critical samples. To address these challenges, we introduce Sparsify, a sparse learning framework specifically designed for Music AVQA. It integrates three sparsification strategies into an end-to-end pipeline and achieves state-of-the-art performance on the Music AVQA datasets. In addition, it reduces training time by 28.32% compared to its fully trained dense counterpart while maintaining accuracy, demonstrating clear efficiency gains. To further improve data efficiency, we propose a key-subset selection algorithm that selects and uses approximately 25% of MUSIC-AVQA v2.0 training data and retains 70-80% of full-data performance across models.