Abstract: Machine translation (MT) has become ubiquitous as a technology that enables individuals to access content on demand in real time that is written in languages they do not speak. However, contrary to recent press releases that have said it has surpassed human quality, the results in practice suggest that it has a long way to go. One of the biggest challenges current-generation neural MT (NMT) faces is that its engines are not easily adaptable and cannot respond to context or extra-linguistic knowledge that human translators routinely deal with. In addition, NMT’s improvements have largely been in terms of fluency (how natural the output sounds) rather than accuracy (how well the translated text represents the content of the source text). This discrepancy in improvement actually increases the risk that critical errors may be remain undetected simply because they are readable and sound plausible.
The next step forward is to build “responsive MT”: systems that can take advantage of embedded metadata about a wide variety of topics and use them to preferentially use the most relevant training data. This metadata includes factors such as text formality, client, product line, text type (e.g., marketing, legal, FAQ, subtitle), intended audience, attributes about the speaker or author, date of authorship, human quality judgments, etc. It also will require systems to look beyond the context of single sentences to include at least the preceding and following segments of text, but also other contexts, such as where in the document a segment occurs and its function (e.g., header, bullet, instruction). This information will help engines return relevant results that are more responsive to their environment and that are more likely to be correct. At the same time, engines will need to know their limits and be able to highlight problems for human attention rather than passing on incorrect, yet plausible-sounding results.
The biggest challenges for developers and implementers alike will be to acquire annotated datasets containing the needed metadata and human quality judgments. Because current datasets generally consist solely of source and target pairs, it will take time for these new approaches to bear fruit. However, organizations such as Microsoft that have experimented with these approaches have seen substantial improvements in quality and reliability of output.
This presentation by Dr. Arle Lommel, senior analyst at CSA Research, will outline the types of metadata that need to be encapsulated and the best practices for gathering them in preparation for the release of responsive MT systems. It will also discuss how these changes are likely to affect technology and translation providers and the new career opportunities that will appear for language professionals.
Bio: Dr. Arle Lommel is a senior analyst with independent market research firm CSA Research. He is a recognized expert in translation quality processes and interoperability standards. Arle’s research focuses on translation technology and the intersection of language and artificial intelligence as well as the value of language in the economy. Born in Alaska, he holds a PhD from Indiana University. Prior to joining CSA Research he worked at the German Research Center for Artificial Intelligence (DFKI) in its Berlin-based language technology lab. In addition to English he speaks fluent Hungarian and passable German, along with bits and pieces of other languages.