通过动态调整训练难度和数据分布，提升音频问答模型的准确率。
Data-Balanced Curriculum Learning for Audio Question Answering
- 用语言模型标注问题难度，按难易程度逐步训练
- 过滤高频音频类别，使数据分布更均衡，准确率提升11.7%
- 引导解码确保答案格式正确，适合需要稳定训练的音频任务
音频问答（AQA）要求模型理解声学内容并进行复杂推理。当前模型在数据集不平衡和训练不稳定方面表现不佳。本文结合课程学习与统计数据平衡策略，利用语言模型标注问题难度，从简单到复杂逐步训练；通过统计过滤移除高频率音频类别，并采用引导解码约束输出为有效多选格式。在DCASE 2025训练集及五个公开数据集上的实验表明，数据筛选使准确率相比基线模型提升11.7%，在DCASE 2025基准上达到64.2%。
原文摘要 · Abstract (English)
Audio question answering (AQA) requires models to understand acoustic content and perform complex reasoning. Current models struggle with dataset imbalances and unstable training dynamics. This work combines curriculum learning with statistical data balancing to address these challenges. The method labels question difficulty using language models, then trains progressively from easy to hard examples. Statistical filtering removes overrepresented audio categories, and guided decoding constrains outputs to valid multiple-choice formats. Experiments on the DCASE 2025 training set and five additional public datasets show that data curation improves accuracy by 11.7% over baseline models, achieving 64.2% on the DCASE 2025 benchmark.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。