AFA-Net: A Differential Attention Approach for Auditory Attention Detection

Abstract

Auditory Attention Detection (AAD) utilizes electroencephalographic (EEG) signals to identify a target speaker in a multi-speaker environment. Despite considerable progress, existing deep learning architectures often lack explicit mechanisms for handling noisy EEG data. To address this limitation, we propose Auditory Focus Attention Networks (AFA-Net), a machine learning framework that replaces vanilla attention with a simple yet flexible differential attention mechanism to help focus on task-relevant neural activity. AFA-Net achieves an upward accuracy of 96.8% at the 2s decision window, while using substantially fewer parameters than most existing methods. To the best of our knowledge, AFA-Net is among the first frameworks to explicitly try to combat EEG noise to improve AAD.

The authors’ abstract, as arXiv lists it. arXiv’s listings, abstracts included, are free for anyone to reuse.

More in Sound

• arXiv preprint · 25 September 2026

  Threat-Aware Energy-Efficient Deployment for Dynamic UAV Networks: A Multi-Agent RL Approach

  Faisal Al-Kamali, Hussein A. Ammar, Francois Chan and 3 others
• arXiv preprint · 26 September 2026

  Schur-Neural KF: Learned Schur-Consistent Corrections to the Extended Kalman Filter

  Min Kim, Lianghao Cao, Soon-Jo Chung and 1 other
• arXiv preprint · 25 September 2026

  Adaptive Pilot Selection for Unified Semantic Communication and Semantic Sensing in ISAC

  Muhammad Abubakar Rashid, Muhammad Hannan Akram, Haejoon Jung and 1 other
• arXiv preprint · 26 September 2026

  Region-Local Copula Evidence Fusion for Heterogeneous Remote Sensing Change Detection

  Zhiyuan Ji, Junjun Yin and Jian Yang
• arXiv preprint · 25 September 2026

  Differential Attention Unlocks Complementary EEG and Speech Fusion for Emotion Recognition

  Philip H. Lee, Shreeram Suresh Chandra and John H. L. Hansen

All new papers in Sound