We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:


Current browse context:


Change to browse by:


References & Citations

DBLP - CS Bibliography


(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Computer Science > Computation and Language

Title: QUAK: A Synthetic Quality Estimation Dataset for Korean-English Neural Machine Translation

Abstract: With the recent advance in neural machine translation demonstrating its importance, research on quality estimation (QE) has been steadily progressing. QE aims to automatically predict the quality of machine translation (MT) output without reference sentences. Despite its high utility in the real world, there remain several limitations concerning manual QE data creation: inevitably incurred non-trivial costs due to the need for translation experts, and issues with data scaling and language expansion. To tackle these limitations, we present QUAK, a Korean-English synthetic QE dataset generated in a fully automatic manner. This consists of three sub-QUAK datasets QUAK-M, QUAK-P, and QUAK-H, produced through three strategies that are relatively free from language constraints. Since each strategy requires no human effort, which facilitates scalability, we scale our data up to 1.58M for QUAK-P, H and 6.58M for QUAK-M. As an experiment, we quantitatively analyze word-level QE results in various ways while performing statistical analysis. Moreover, we show that datasets scaled in an efficient way also contribute to performance improvements by observing meaningful performance gains in QUAK-M, P when adding data up to 1.58M.
Subjects: Computation and Language (cs.CL)
Journal reference: COLING 2022
Cite as: arXiv:2209.15285 [cs.CL]
  (or arXiv:2209.15285v2 [cs.CL] for this version)

Submission history

From: Sugyeong Eo [view email]
[v1] Fri, 30 Sep 2022 07:47:44 GMT (638kb,D)
[v2] Tue, 29 Nov 2022 12:05:07 GMT (637kb,D)

Link back to: arXiv, form interface, contact.