AudSemThinker: Enhancing Audio-Language Models Through Reasoning over Semantics of Sound
Gijs Wijngaard, Elia Formisano, Michele Esposito, Michel Dumontier
Abstract
Audio-language models have shown promising results in various sound understanding tasks, yet they remain limited in their ability to reason over the fine-grained semantics of sound. In this paper, we present AudSemThinker, a model whose reasoning is structured around a framework of auditory semantics inspired by human cognition. To support this, we introduce AudSem, a novel dataset specifically curated for semantic descriptor reasoning in audio-language models. AudSem addresses the persistent challenge of data contamination in zero-shot evaluations by providing a carefully filtered collection of audio samples paired with captions generated through a robust multi-stage pipeline. Our experiments demonstrate that AudSemThinker outperforms state-of-the-art models across multiple training settings, highlighting its strength in semantic audio reasoning. Both AudSemThinker and the AudSem dataset are released publicly.
BibTeX
@inproceedings{
wijngaard2025audsemthinker,
title={AudSemThinker: Enhancing Audio-Language Models Through Reasoning over Semantics of Sound},
author={Gijs Wijngaard and Elia Formisano and Michele Esposito and Michel Dumontier},
booktitle={The Thirty-ninth Annual Conference on Neural Information Processing Systems},
year={2025},
url={https://openreview.net/forum?id=pozsP0ZcZN}
}