SoundMind: RL-Incentivized Logic Reasoning for Audio-Language Models
Abstract
While large language models have demonstrated impressive reasoning abilities, their extension to the audio modality, particularly within large audio-language models (LALMs), remains underexplored. Addressing this gap requires a systematic approach that involves a capable base model, high-quality reasoning-oriented audio data, and effective training algorithms. In this work, we present a comprehensive solution for audio logical reasoning (ALR) tasks: we introduce SoundMind, a dataset of 6,446 audio-text annotated samples specifically curated to support complex reasoning. Building on this resource, we propose SoundMind-RL, a rule-based reinforcement learning (RL) algorithm designed to equip audio-language models with robust audio-text reasoning capabilities. By fine-tuning Qwen2.5-Omni-7B on the proposed SoundMind dataset using SoundMind-RL, we achieve strong and consistent improvements over state-of-the-art baselines on the SoundMind benchmark. This work highlights the benefit of combining high-quality, reasoning-focused datasets with specialized RL techniques, and contributes to advancing auditory intelligence in language models. The code and dataset introduced in this work are publicly available at https://github.com/xid32/SoundMind.
Cite
@article{arxiv.2506.12935,
title = {SoundMind: RL-Incentivized Logic Reasoning for Audio-Language Models},
author = {Xingjian Diao and Chunhui Zhang and Keyi Kong and Weiyi Wu and Chiyu Ma and Zhongyu Ouyang and Peijun Qing and Soroush Vosoughi and Jiang Gui},
journal= {arXiv preprint arXiv:2506.12935},
year = {2025}
}
Comments
Accepted to EMNLP 2025 Main Conference (Oral Presentation)