SAKURA: On the Multi-hop Reasoning of Large Audio-Language Models Based on Speech and Audio Information.

Chih-Kai Yang, Neo Ho, Yen-Ting Piao, Hung-yi Lee

Browse the full Interspeech paper archive.