作者位次 #4/6ACL 2026Findings· Findings of ACL
AUDITA : A New Dataset to Audit Humans vs. AI Skill at Audio QA
Tasnim Kabir · Dmytro Kurdydyk · Aadi Palnitkar · Liam Dorn · Ahmed Haj Ahmed · Jordan Lee Boyd-Graber
摘要
Existing audio question answering benchmarks largely emphasize sound event classification or caption-grounded queries, often enabling models to succeed through shortcut strategies, short-duration cues, lexical priors, dataset-specific biases, or even bypassing audio via metadata and captions rath...