Automatic annotation of short-text data to a large number of target labels, referred to as Short Text Extreme Classification, has found numerous applications including prediction of related searches and product recommendation tasks. In this paper, we propose a convolutional architecture InceptionXML which is light-weight, yet powerful, and robust to the inherent lack of word-order in short-text queries encountered in search and recommendation tasks. We demonstrate the efficacy of applying convolutions by recasting the operation along the embedding dimension instead of the word dimension as applied in conventional CNNs for text classification. Towards scaling our model to datasets with millions of labels, we also propose InceptionXML+ framework which improves upon the shortcomings of the recently proposed dynamic hard-negative mining technique for label shortlisting by synchronizing the label-shortlister and extreme classifier. InceptionXML+ not only reduces the inference time to half but is also an order of magnitude smaller than previous state-of-the-art Astec in terms of model size. Through our proposed models, we outperform all existing approaches on popular benchmark datasets.

翻译：短文本数据到大量目标标签的自动标注，即短文本极端分类，已在相关搜索预测与产品推荐等任务中得到广泛应用。本文提出一种轻量而高效的卷积架构InceptionXML，该架构对搜索与推荐场景中短文本查询固有的词序缺失问题具有鲁棒性。通过将卷积操作重新定义在嵌入维度而非传统文本分类CNN所使用的词维度上，我们论证了卷积在此任务中的有效性。为将模型扩展至数百万标签的数据集，我们进一步提出InceptionXML+框架，该框架通过同步标签筛选器与极端分类器，改进了近期提出的动态硬负挖掘技术在标签短列表生成中的缺陷。与现有最优模型Astec相比，InceptionXML+不仅将推理时间减半，模型规模更缩小一个数量级。基于所提模型，我们在多个权威基准数据集上全面超越现有方法。