SAO WMT19 Test Suite: Machine Translation of Audit Reports

TL;DR

This paper introduces a specialized machine translation test set for audit reports, evaluating the performance of general MT systems in the auditing domain and highlighting their limitations in preserving domain-specific semantics.

Contribution

It provides a new domain-specific test suite for MT evaluation and analyzes the performance gap of general MT systems on audit report translation.

Findings

General MT systems perform well on audit reports but lack deep domain understanding.

Automatic evaluation metrics are insufficient for capturing domain-specific translation quality.

Even top systems fail to preserve critical semantic details in legal documents.

Abstract

This paper describes a machine translation test set of documents from the auditing domain and its use as one of the "test suites" in the WMT19 News Translation Task for translation directions involving Czech, English and German. Our evaluation suggests that current MT systems optimized for the general news domain can perform quite well even in the particular domain of audit reports. The detailed manual evaluation however indicates that deep factual knowledge of the domain is necessary. For the naked eye of a non-expert, translations by many systems seem almost perfect and automatic MT evaluation with one reference is practically useless for considering these details. Furthermore, we show on a sample document from the domain of agreements that even the best systems completely fail in preserving the semantics of the agreement, namely the identity of the parties.