用专家模型与原型分类器实现音频领域增量学习，避免遗忘且无需存储原始数据。
Domain-incremental audio classification using domain-specific experts and prototype classifier
- 冻结专家模型+原型分类器，通过特征拼接防止灾难性遗忘。
- 采用生成回放和跨阶段插补，不存音频也能保留历史知识。
- 在DCASE 2026挑战中达到78.15%微准确率，优于单个模型。
本文提交了针对DCASE 2026挑战赛任务7（领域增量音频分类）的系统方案。核心难点在于系统无法一次性访问过去或未来领域的数据。我们将领域增量学习（DIL）视为一个固定特征回放问题：在每个增量阶段，训练一个或两个紧凑专家模型并保持冻结；最终阶段，将所有冻结专家的倒数第二层特征拼接后，仅用缓存特征训练一个轻量级每类原型分类器。该设计通过保留专家模型避免灾难性遗忘。为在不存储原始音频的前提下保留早期领域知识，部分专家采用基于DeepInversion的生成回放进行训练。同时训练了一个跨阶段回归插补器，填补早期阶段尚未存在的专家特征空缺。共提交四个完全符合DIL要求的系统：三个基于不同冻结五专家主干的系统及其跨堆栈集成，在开发集上取得78.15%微平均准确率和77.03%宏平均准确率，优于任一单个主干模型。
原文摘要 · Abstract (English)
This technical report presents submission systems for Task 7(domain-incremental audio classification) of the DCASE 2026 Challenge. The main obstacle is that, the system is unable to access to past or future domain's data at once. We approached domain-incremental learning (DIL) as a frozen-feature replay problem. At each incremental stage, one or two compact experts are trained and then kept fixed; at the final stage, the penultimate features from all frozen experts are concatenated and used to train a lightweight per-class prototype classifier solely on cached features. This design prevents catastrophic forgetting by preserving each expert models at inference. To retain earlier-domain knowledge without storing raw audio, some experts were trained with DeepInversion-based generative replay. A cross-stage regression imputer was trained to fill the expert feature slots that did not yet exist at an ealier stage. We submit four fully DIL-compliant systems: three systems based on diverse frozen five-expert backbones and their cross-stack ensemble achieving 78.15% micro / 77.03% macro on the development set, outperforming every individual backbone on both evaluations.
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。