NewsEmbed: Modeling News through Pre-trained Document Representations
This work addresses the problem of scaling document-level models for news content, which is incremental as it builds on pre-trained representations and contrastive learning techniques.
The authors tackled the challenge of modeling news articles at the document-level by proposing a method to mine billions of high-quality training examples with minimal human supervision, resulting in a universal document encoder that shows competitive performance on multiple natural language understanding tasks.
Effectively modeling text-rich fresh content such as news articles at document-level is a challenging problem. To ensure a content-based model generalize well to a broad range of applications, it is critical to have a training dataset that is large beyond the scale of human labels while achieving desired quality. In this work, we address those two challenges by proposing a novel approach to mine semantically-relevant fresh documents, and their topic labels, with little human supervision. Meanwhile, we design a multitask model called NewsEmbed that alternatively trains a contrastive learning with a multi-label classification to derive a universal document encoder. We show that the proposed approach can provide billions of high quality organic training examples and can be naturally extended to multilingual setting where texts in different languages are encoded in the same semantic space. We experimentally demonstrate NewsEmbed's competitive performance across multiple natural language understanding tasks, both supervised and unsupervised.