让音频模型听声时‘思考’，提升分类准确率。
Thinking While Listening: Simple Test Time Scaling For Audio Classification
- 在听音时引入推理机制，增强类别空间的思考能力。
- 测试阶段增加采样次数，准确率持续提升。
- 轻量重训效果超大语言模型，适合资源有限场景。
我们提出一种框架，使神经网络在聆听日常声音时具备‘思考’能力，从而提升音频分类性能。受大型语言模型推理能力的启发，本文探讨两个核心问题：(i) 如何将推理融入现有音频分类流程，在类别空间中实现推理以提升性能；(ii) 能否从零设计新架构，同时支持思考与测试阶段缩放？实验表明，两种设定下模型均表现出更高的分类准确率。利用测试阶段缩放，随着采样轨迹数量增加，性能持续提升。此外，评估了两款开源推理模型（GPT-OSS-20B 和 Qwen3-14B），发现尽管它们具备零样本推理能力，但仅重训小模型（如 GPT-2）的嵌入层，即可超越数十亿参数的文本推理模型表现。
原文摘要 · Abstract (English)
We propose a framework that enables neural models to "think while listening" to everyday sounds, thereby enhancing audio classification performance. Motivated by recent advances in the reasoning capabilities of large language models, we address two central questions: (i) how can thinking be incorporated into existing audio classification pipelines to enable reasoning in the category space and improve performance, and (ii) can a new architecture be designed from the ground up to support both thinking and test-time scaling? We demonstrate that in both settings, our models exhibit improved classification accuracy. Leveraging test-time scaling, we observe consistent gains as the number of sampled traces increases. Furthermore, we evaluate two open-source reasoning models, GPT-OSS-20B and Qwen3-14B, showing that while such models are capable of zero-shot reasoning, a lightweight approach--retraining only the embedding matrix of a frozen, smaller model like GPT-2--can surpass the performance of billion-parameter text-based reasoning models.
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。