Researchers have introduced AIB, the first benchmark designed to evaluate Large Audio Language Models (LALMs) on their ability to replicate human auditory illusions. The benchmark covers ten distinct illusions across music, sound, and speech, incorporating knowledge-based priors. While current LALMs tend to be faithful to the raw audio signal in simpler illusions, some models show more human-like responses when linguistic or musical context is involved, though none fully match human perception. This work aims to provide a new method for understanding the cognitive capabilities of LALMs. AI
IMPACT This benchmark could reveal limitations in AI's understanding of complex auditory perception, guiding future model development.
RANK_REASON The cluster contains an academic paper introducing a new benchmark for AI models. [lever_c_demoted from research: ic=1 ai=1.0]
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →
