用生成描述的方式做多标签分类，效果更好更通用。
Label-semantics Aware Generative Approach for Domain-Agnostic Multilabel Classification
- 把标签当作可生成的描述，输入文本生成对应标签描述。
- 在多个数据集上提升13.94%（Micro-F1）和24.85%（Macro-F1）。
- 模型轻量通用，适合实际部署，尤其适合跨领域场景。
文本数据爆炸使得手动文档分类愈发困难。为此，我们提出一种鲁棒、高效的无领域限制生成式多标签文本分类框架。不同于将标签视为简单符号，本方法利用预定义的标签描述，并训练模型根据输入文本生成这些描述。推理时，通过微调的句子嵌入模型将生成的描述与预定义标签匹配。我们结合交叉熵损失与生成句向量和目标描述间的余弦相似度，构建双目标损失函数，确保语义对齐与分类准确。所提模型 LAGAMC 在参数效率与跨数据集泛化能力上表现优异，适用于实际应用。我们在所有评测数据集上均达到新基准性能，相比最接近的基线模型，平均提升13.94%（Micro-F1）和24.85%（Macro-F1）。
原文摘要 · Abstract (English)
The explosion of textual data has made manual document classification increasingly challenging. To address this, we introduce a robust, efficient domain-agnostic generative model framework for multi-label text classification. Instead of treating labels as mere atomic symbols, our approach utilizes predefined label descriptions and is trained to generate these descriptions based on the input text. During inference, the generated descriptions are matched to the pre-defined labels using a finetuned sentence transformer. We integrate this with a dual-objective loss function, combining cross-entropy loss and cosine similarity of the generated sentences with the predefined target descriptions, ensuring both semantic alignment and accuracy. Our proposed model LAGAMC stands out for its parameter efficiency and versatility across diverse datasets, making it well-suited for practical applications. We demonstrate the effectiveness of our proposed model by achieving new state-of-the-art performances across all evaluated datasets, surpassing several strong baselines. We achieve improvements of 13.94% in Micro-F1 and 24.85% in Macro-F1 compared to the closest baseline across all datasets.
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。