Our new X account is live! Follow @wizwand_team for updates

Audio-Visual Question Answering on Music-AVQA 2000 samples

13.8ASR Rate

Combined Loss

Updated 4d ago

Evaluation Results

Method	Links
Combined Loss 2026.01		13.8
Encoder-Based Cosine Similarity Loss 2026.01		12.92
Negative Language Modeling Loss 2026.01		10.74
Hidden-State Similarity Loss 2026.01		10.28
Audio Attention Amplification Loss 2026.01		9.81
Attention Randomization Loss 2026.01		9.81
Vision Attention Suppression Loss 2026.01		7.94