Audio-R1：基于自监督任务强化学习的音频推理框架
Audio-R1: Audio Reasoning through Reinforcement Learning on Self-Supervised Tasks
AI 导读
研究者提出 Audio-R1，一个强化学习后训练框架，把音频自监督学习作为可验证奖励来源，提升大音频语言模型（LALM）的音频理解与推理能力。该框架利用自监督任务构建时自动确定的答案验证模型响应并计算奖励，在多个 LALM 和多种音频基准上取得一致提升，消融实验验证了各任务收益及联合训练的额外增益。相比监督微调，Audio-R1 还能提升与训练任务不同的下游任务表现和推理轨迹质量。
正文 · 原文
Abstract:Reasoning is increasingly important for large audio-language models (LALMs), yet remains relatively underexplored compared with text-only and vision-language models. We introduce Audio-R1, a reinforcement learning (RL) post-training framework that revisits audio self-supervised learning as a source of verifiable rewards for improving audio understanding and reasoning task performance. Audio-R1 uses answers automatically determined during the construction of self-supervised tasks to verify model responses and compute rewards. Experiments across multiple LALMs and diverse audio benchmarks show consistent performance gains, while ablations demonstrate the benefits of each self-supervised task and additional gains from joint training. While both supervised fine-tuning and Audio-R1 improve self-supervised task accuracy, Audio-R1 further improves performance on downstream tasks distinct from the training tasks. These results highlight the potential of RL post-training with self-supervised audio tasks to improve downstream performance and reasoning-trace quality.
| Subjects: | Sound (cs.SD); Artificial Intelligence (cs.AI); Audio and Speech Processing (eess.AS) |
| Cite as: | arXiv:2610.34609 [cs.SD] |
| (or arXiv:2610.34609v1 [cs.SD] for this version) | |
| https://doi.org/10.48550/arXiv.2610.34609 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Cem Subakan [view email]
[v1]
Sun, 25 Oct 2026 22:17:04 UTC (343 KB)
来源：arXiv cs.AI · arxiv.org