Work

Findings of the WMT 2018 Shared Task on Parallel Corpus Filtering

Year: 2018

Type: conference-paper

Abstract: We posed the shared task of assigning sentence-level quality scores for a very noisy corpus of sentence pairs crawled from the web, with the goal of sub-selecting 1% and 10% of high-quality data to be used to train machine translation systems.Seventeen participants from companies, national research labs, and universities participated in this task.

DOI: 10.18653/v1/w18-6453

Authors: Philipp Koehn, Huda Khayrallah, Kenneth Heafield, Mikel L. Forcada

Institutions: Johns Hopkins University, University of Edinburgh, University of Alicante

Language: English

Topic: Natural Language Processing Techniques

Subfield: Artificial Intelligence

Field: Computer Science

Domain: Physical Sciences

Open Access status: gold

Funders: Johns Hopkins University, European Commission

OpenAlex ID: W2902918014

FWCI: 7.627

Cites: 48

Cited by: 99

Related to: 10