A meta-benchmark that auto-generates multimodal music-perception multiple-choice tests from user symbolic music, demonstrated on ChoraleBricks with text-only and white-noise controls.
arXiv preprint arXiv:2509.15662 , year=
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
cs.SD 2years
2026 2verdicts
UNVERDICTED 2representative citing papers
GST uses gradient-based affinity metrics to form dataset groups and applies progressive scheduling, achieving 30-40% faster convergence than uniform mixture training on 14 AudioQA datasets while matching or exceeding performance.
citing papers explorer
-
Music I Care About: Automated Multimodal Benchmarking of LLM Music Perception Skills on (Almost) Any Music
A meta-benchmark that auto-generates multimodal music-perception multiple-choice tests from user symbolic music, demonstrated on ChoraleBricks with text-only and white-noise controls.
-
Heterogeneity-Aware Dataset Scheduling for Efficient Audio Large Language Model Training
GST uses gradient-based affinity metrics to form dataset groups and applies progressive scheduling, achieving 30-40% faster convergence than uniform mixture training on 14 AudioQA datasets while matching or exceeding performance.