AURA Score, a new metric for Audio Question Answering, achieved the highest correlation with human judgments on the AQEval benchmark of 10,000 model‑generated answers. https://getnews.me/aura-score-new-metric-for-evaluating-audio-question-answering/ #aura #aqeval #audioqa

Omni-CLST reaches 73.80% accuracy on MMAU-mini and a new state‑of‑the‑art 64.30% on MMAR, using error‑aware curriculum learning and selective chain‑of‑thought. Read more: https://getnews.me/omni-clst-advances-audio-qa-with-error-aware-curriculum-learning/ #audioqa #curriculumlearning

Omni-CLST, an error‑aware curriculum framework for audio QA, hit 73.8% on MMAU‑mini and set a 64.3% record on MMAR. Read more: https://getnews.me/omni-clst-improves-audio-question-answering-with-curriculum-learning/ #audioqa #curriculumlearning