What We’re Reading: Single-Queue Decoding for Neural Machine Translation

The most popular way of finding a translation for a source sentence with a neural sequence-to-sequence model is a simple beam search. The target sentence is predicted one word at a time and after each prediction, a fixed number of possibilities (typically between 4 and 10) is retained for further exploration. This strategy can be… Read More What We’re Reading: Single-Queue Decoding for Neural Machine Translation

What We’re Reading: Neural Machine Translation with Reconstruction

Neural MT systems generate translations one word at a time. They can still generate fluid translations because they choose each word based on all of the words generated so far. Typically, these systems are just trained to generate the next word correctly, based on all previous words. One systematic problem with this word-by-word approach to training and… Read More What We’re Reading: Neural Machine Translation with Reconstruction

Webinar: How to Run a Data-Driven MT Evaluation

There are many machine translation systems on the market today. Choosing the most effective system for a particular production workflow can be a difficult task. Lilt’s CEO and co-founder, Spence Green, recently hosted a webinar to show a principled, data-driven method that is used by researchers to score systems. The webinar covered automatic measures such… Read More Webinar: How to Run a Data-Driven MT Evaluation

2017 Machine Translation Quality Evaluation Addendum

This post is an addendum to our original post on 1/10/2017 entitled 2017 Machine Translation Quality Evaluation. Experimental Design We evaluate all machine translation systems for English-French and English-German. We report case-insensitive BLEU-4 [2], which is computed by the mteval scoring script from the Stanford University open source toolkit Phrasal (https://github.com/stanfordnlp/phrasal). NIST tokenization was applied… Read More 2017 Machine Translation Quality Evaluation Addendum

2017 Machine Translation Quality Evaluation

The language services industry now has an array of machine translation options. The goal of Lilt Labs is to provide reproducible and unbiased evaluations of these options using public datasets and a rigorous methodology. The 2017 evaluation is intended to assess machine translation quality in a prototypical translation workflow. Therefore, it includes not only an evaluation of baseline translation… Read More 2017 Machine Translation Quality Evaluation

Technology for Interactive MT

This article describes the technology behind Lilt’s interactive translation suggestions. The details were first published in an academic conference paper, Models and Inference for Prefix-Constrained Machine Translation. Machine translation systems can translate whole sentences or documents, but they can also be used to finish translations that were started by a person — a form of… Read More Technology for Interactive MT

Morphing into the Promised Land

Guest post by Jost Zetzsche, originally published in Issue 16-12-268 of The Tool Box Journal. Some of you know that I’ve been very interested in morphology. No, let me put that differently: I’ve been very frustrated that the translation environment tools we use don’t offer morphology. There are some exceptions — such as SmartCat, Star Transit, Across,… Read More Morphing into the Promised Land

Case Study: First Large-Scale Application of Auto-Adaptive MT

Combining Machine Translation (MT) with auto-adaptive Machine Learning (ML) enables a new paradigm of machine assistance. Such systems learn from the experience, intelligence and insights of their human users, improving productivity by working in partnership, making suggestions and improving accuracy over time. The net result is that human reviewers produce far higher volumes of content,… Read More Case Study: First Large-Scale Application of Auto-Adaptive MT

Interactive and Adaptive Computer Aided Translation

Originally published on Kirti Vashee’s blog eMpTy Pages. Lilt is an interactive and adaptive computer-aided translation tool that integrates machine translation, translation memories, and termbases into one interface that learns from translators. Using Lilt is an entirely different experience from post-editing machine translations — an experience that our users love, and one that yields substantial productivity… Read More Interactive and Adaptive Computer Aided Translation

Case Study: SDL Trados

Abstract: We compare human translation performance in Lilt to SDL Trados, a widely used computer-aided translation tool. Lilt generates suggestions via an adaptive machine translation system, whereas SDL Trados relies primarily on translation memory. Five in-house English–French translators worked with each tool for an hour. Client data for two genres was translated. For user interface… Read More Case Study: SDL Trados