Reference Bias in Monolingual Machine Translation Evaluation

  title={Reference Bias in Monolingual Machine Translation Evaluation},
  author={Marina Fomicheva and Lucia Specia},
In the translation industry, human translations are assessed by comparison with the source texts. In the Machine Translation (MT) research community, however, it is a common practice to perform quality assessment using a reference translation instead of the source text. In this paper we show that this practice has a serious issue – annotators are strongly biased by the reference translation provided, and this can have a negative impact on the assessment of MT quality. 

From This Paper

Figures, tables, and topics from this paper.


Publications referenced by this paper.
Showing 1-10 of 16 references

Official Results of the NIST 2008 “Metrics for MAchine TRanslation” Challenge (MetricsMATR08)

  • Mark Przybocki, Kay Peterson, Sebastian Bronsart.
  • Proceedings of the AMTA-2008 Workshop on Metrics…
  • 2008
3 Excerpts

Similar Papers

Loading similar papers…