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

Download:

Current browse context:

cs.CL

Change to browse by:

cs

References & Citations

DBLP - CS Bibliography

Bookmark

(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: Input Augmentation Improves Constrained Beam Search for Neural Machine Translation: NTT at WAT 2021

Abstract: This paper describes our systems that were submitted to the restricted translation task at WAT 2021. In this task, the systems are required to output translated sentences that contain all given word constraints. Our system combined input augmentation and constrained beam search algorithms. Through experiments, we found that this combination significantly improves translation accuracy and can save inference time while containing all the constraints in the output. For both En->Ja and Ja->En, our systems obtained the best evaluation performances in automatic evaluation.
Comments: 9 pages, 4 figures, WAT 2021 Restricted Translation Task
Subjects: Computation and Language (cs.CL)
Cite as: arXiv:2106.05450 [cs.CL]
  (or arXiv:2106.05450v1 [cs.CL] for this version)

Submission history

From: Katsuki Chousa [view email]
[v1] Thu, 10 Jun 2021 01:39:59 GMT (2391kb,D)

Link back to: arXiv, form interface, contact.