DeepAI AI Chat
Log In Sign Up

Weakly Supervised Grammatical Error Correction using Iterative Decoding

by   Jared Lichtarge, et al.

We describe an approach to Grammatical Error Correction (GEC) that is effective at making use of models trained on large amounts of weakly supervised bitext. We train the Transformer sequence-to-sequence model on 4B tokens of Wikipedia revisions and employ an iterative decoding strategy that is tailored to the loosely-supervised nature of the Wikipedia training corpus. Finetuning on the Lang-8 corpus and ensembling yields an F0.5 of 58.3 on the CoNLL'14 benchmark and a GLEU of 62.4 on JFLEG. The combination of weakly supervised training and iterative decoding obtains an F0.5 of 48.2 on CoNLL'14 even without using any labeled GEC data.


page 1

page 2

page 3

page 4


Corpora Generation for Grammatical Error Correction

Grammatical Error Correction (GEC) has been recently modeled using the s...

Controllable Data Synthesis Method for Grammatical Error Correction

Due to the lack of parallel data in current Grammatical Error Correction...

Parallel Iterative Edit Models for Local Sequence Transduction

We present a Parallel Iterative Edit (PIE) model for the problem of loca...

Grammatical Error Correction as GAN-like Sequence Labeling

In Grammatical Error Correction (GEC), sequence labeling models enjoy fa...

Converge to the Truth: Factual Error Correction via Iterative Constrained Editing

Given a possibly false claim sentence, how can we automatically correct ...

Hard Mixtures of Experts for Large Scale Weakly Supervised Vision

Training convolutional networks (CNN's) that fit on a single GPU with mi...

Training ASR models by Generation of Contextual Information

Supervised ASR models have reached unprecedented levels of accuracy, tha...