Abstract
<jats:title>Abstract</jats:title> <jats:p>This article discusses post‐editing of machine translation (PEMT), in which a translator compares a source text with a translation produced by machine translation (MT) and revises it to make it acceptable for its intended purpose. The history of MT is briefly discussed; it has evolved from rule‐based, to statistical, to the current neural MT and is increasingly used in the translation industry to reduce effort and costs. Two major modes of post‐editing are described, static and interactive, along with the differences in translator agency in the two modes. Static post‐editing takes a complete text from MT and revises it to a given standard of quality (full or light), while interactive post‐editing occurs when a translator is crafting a translation and can use or reject MT suggestions. The types and distribution of translation accuracy and language fluency errors are discussed, which differ between MT and human translation (HT). Finally, current research trends in post‐editing are described. Research in post‐editing has increased significantly since 2015.</jats:p>