Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
This paper introduces the Auditory Illusion Benchmark (AIB), the first dedicated benchmark for evaluating Large Audio Language Models (LALMs) against auditory perceptual illusions. By comparing model responses to controlled human listening studies across ten auditory illusions, the research reveals that while LALMs maintain fidelity to low-level acoustic signals, they only partially replicate human-like responses when linguistic or musical context is involved. The findings underscore the limitations of LALMs as cognitive models and establish a new framework for exploring auditory cognition in AI systems.
LALMs exhibit surprising biases in auditory perception, showing human-like responses only in the presence of linguistic or musical context.
Perceptual illusions have long served as crucial probes into human cognition, revealing biases and limitations of perception. In the auditory domain, such illusions provide a unique lens for testing whether Large Audio Language Models (LALMs) replicate human perceptual tendencies. Despite their importance, most benchmarks focus on visual illusions or general audio tasks, leaving auditory illusions underexplored. To this end, we present AIB, the first auditory illusion benchmark for LALMs, covering ten representative illusions across music, sound, and speech, each annotated for the presence of knowledge-based priors. Our methodology pairs model evaluation with controlled human listening studies, enabling direct comparison of responses. Results show systematic differences: while most LALMs remain signal-faithful on low-level acoustic illusions, several exhibit more human-like responses when linguistic or musical priors are involved, although no model matches the human perceptual profile. These findings highlight the current limitations of LALMs as cognitive models. By establishing auditory illusions as a rigorous testbed, our work offers a new perspective for probing neural black-box models and advancing understanding of auditory cognition. AIB is publicly available at https://github.com/gillosae/aib.