Phrase-based Machine Translation is State-of-the-Art for Automatic Grammatical Error Correction

2016/05/20 by Junczys-Dowmunt, Marcin, Grundkiewicz, Roman · 2 citations
#Computation and Language (cs.CL) #FOS: Computer and information sciences

paper · doi:10.48550/arxiv.1605.06353

Abstract

In this work, we study parameter tuning towards the M² metric, the standard metric for automatic grammar error correction (GEC) tasks. After implementing M² as a scorer in the Moses tuning framework, we investigate interactions of dense and sparse features, different optimizers, and tuning strategies for the CoNLL-2014 shared task. We notice erratic behavior when optimizing sparse feature weights with M² and offer partial solutions. We find that a bare-bones phrase-based SMT setup with task-specific parameter-tuning outperforms all previously published results for the CoNLL-2014 test set by a large margin (46.37% M² over previously 41.75%, by an SMT system with neural features) while being trained on the same, publicly available data. Our newly introduced dense and sparse features widen that gap, and we improve the state-of-the-art to 49.49% M².

Cited by

• Time Is Effort: Estimating Human Post-Editing Time for Grammar Error Correction Tool Evaluation
• Opportunities and Challenges of LLMs in Education: An NLP Perspective

Related

• Near Human-Level Performance in Grammatical Error Correction with Hybrid Machine Translation
• Data Weighted Training Strategies for Grammatical Error Correction
• Improving Grammatical Error Correction with Machine Translation Pairs
• Approaching Neural Grammatical Error Correction as a Low-Resource\n Machine Translation Task
• Towards Unsupervised Grammatical Error Correction using Statistical Machine Translation with Synthetic Comparable Corpus