Corruption Is Not All Bad: Incorporating Discourse Structure into Pre-training via Corruption for Essay Scoring

Abstract

Existing approaches for automated essay scoring and document representation learning typically rely on discourse parsers to incorporate discourse structure into text representation. However, the performance of parsers is not always adequate, especially when they are used on noisy texts, such as student essays. In this paper, we propose an unsupervised pre-training approach to capture discourse structure of essays in terms of coherence and cohesion that does not require any discourse parser or annotation. We introduce several types of token, sentence and paragraph-level corruption techniques for our proposed pre-training approach and augment masked language modeling pre-training with our pre-training method to leverage both contextualized and discourse information. Our proposed unsupervised approach achieves new state-of-the-art result on essay Organization scoring task.

References in corpus (9)

• Distributed Representations of Sentences and Documents
• Longformer: The Long-Document Transformer
• DocBERT: BERT for Document Classification
• HIBERT: Document Level Pre-training of Hierarchical Bidirectional Transformers for Document Summarization
• Automated Essay Scoring based on Two-Stage Learning
• Language Model Pre-training for Hierarchical Document Representations
• Word Mover's Embedding: From Word2Vec to Document Embedding
• Neural Multi-task Learning in Automated Assessment
• Neural Discourse Structure for Text Categorization