The MTPE Training Protocol - Interview with Viveta Gene (Intertranslation) and Lucía Guerrero (CPSL)

The machine translation post-editing (MTPE) Training Special Interest Group (SIG) operates as a time-limited, collaborative community project. With monthly teleconferences and offline work, the SIG is developing a set of best practices for training professionals who will conduct post-editing activities.

Watch (or read!) our interview with Viveta Gene and Lucía Guerrero, the moderators of the GALA MTPE SIG.

For more from GALA Experts, subscribe to our YouTube Channel

The work of the MTPE Special Interest Group at GALA

Viveta Gene (VG): MTPE special interest Group is an initiative born after my webinars for GALA – The Management and Training Challenges of MTPE which revealed the need for training not only for post-editors but for all stakeholders, LSPs, Academia and Clients. So, I am really thankful to GALA for embracing this initiative and organizing the MTPE Training SIG and to Lucia Guerrero, my co-moderator for believing in the vision of this SIG and working with me, it’s been more than one year. Now, in this scope, the SIG could be considered as a Common Intelligence Workshop where we want all the voices to be heard and all the groups to be represented. To do so, we have appointed one representative for each group, Post-Editors, Academia, Clients and LSPs to moderate the groups and keep the minutes. We organize interviews with the members of the group to shed light on the industrial workflows and practices, share the perspectives of the post-editors, understand the challenges of the clients…And we do not limit ourselves to the monthly calls.

Lucia Guerrero (LG): Apart from the monthly calls we also have a place where we can continue the conversation and this place is Basecamp. Basecamp is an online platform where all the SIG members can connect to post questions, topics for discussion and comment anyone else’s topics. Viveta and I also upload the presentations and the minutes, and any other reference material related to our monthly calls such as academic papers. Sometimes 1.5 h is not enough and unfortunately, we don’t have the time to listen to everyone, so in this sense Basecamp has proved to be really effective. It’s also helped us to bring new ideas and points of view to the calls. And it also creates this sense of community: it’s not just: let’s meet once monthly and then let’s forget about each other until next month. Basecamp helps us feel that we are part of a group where we all collaborate towards an objective and adds continuity.

VG: With the contribution of all the members, our vision is by September to draft and produce a Common Machine Translation Post-Editing Training Protocol which will represent the current state of MTPE in our industry with the perspectives of all groups and will shape the MTPE service and training.

For more on the GALA Special Interest Groups

Post-editing vs. translation: Are they really that different?

VG: For the MTPE Training SIG, post-editing is not a different task from translation. It is translation, in the sense that it communicates the message of the text from the source language to the target language from the author to the target audience. Post-Editing is based on the translation process in terms of cognitive background and skills; however, it needs an extensive knowledge and use of technology and the upgrading and reskilling of the linguists.

LG: As Viveta said, based in our discussions on the SIG, maybe we can say that post-editing is not exactly a task different from translation, but maybe a type of translation task. In the past, especially with SMT, some people supported the idea that post-editing was just a matter of swapping words here and there, and that, to post-edit, you didn’t need to be proficient in the source language (I’m not supporting this, I’m just saying that it was a common conception). This conception has changed a lot, especially since the use of NMT, which is prevalent now. In the SIG we’ve seen that now it is widely accepted that, in order to post-edit, you really need to be proficient not only in your target language, but also in your source language, and this means that you need translation skills first plus other skills.

Developing skills for post-editing

VG: The technicality of the task reveals the need of development of technical skills and some additional ones, like problem-solving and decision-making. Based on the SIG discussion, experience in CAT tools and translation of more than 3 years seems to work well for the profile of the post-editor in terms of performance, quality and speed.

LG: Apart from the skills, we think we must also speak about the competences. For example, instrumental competences (those related to the tools, i.e. some basic information about how the MT system works) and strategic competence (being able to make quick decisions such as: which is going to be faster: fixing the machine-translated segment, or deleting it and translate from scratch?), and finally attitudinal competence (research has proved that a negative attitude towards MT can influence the whole post-editing experience, and of course this doesn’t mean that we must accept bad quality MT, but if we start a post-editing task with negative preconceptions it is more likely that we end up deleting the MT suggestions and translating from scratch even in those cases when the proposals were appropriate). I think this applies not only to MT but to review of human translation: some people tend to apply much more changes than needed, and that is an attitude.

The state of MTPE training in academia and in the industry

LG: I think one the main takeaway from our monthly calls is that we must accept that academia cannot absorb the whole burden of training post-editors. It is simply impossible. First, the variety of use cases is huge, and the amount and type of corrections depend on too many factors, such as the purpose of the translation and the target audience, which then define the quality requirements, which in turn define the precise aspects need to be fixed or not. So, in the SIG we agreed that LSPs and customers, that is, the translation requesters should also take the responsibility of training post-editors and prepare onboarding and training processes and feedback loops which add something else to the academia curriculum, for example giving information about customer-specific needs and particular workflows.

VG: We could say that machine translation post-editing training is now start making the first steps. For Academia, it is included in some Master Programmes; however, this depends on the country as for some countries still talking about translators using MT is not acceptable. We also have individual trainers and training courses provided by LSPs and organizations like TAUS. The training provided by the Academia focuses more on the theory of the MT, the errors and not on the development of the specific skills that differentiate translation from post-editing. The training of the LSPs focuses on the speed of the MT with gaps in the definitions of the service and the how-to, sometimes being generic, not presenting the errors of the specific engine and not specifying the quality that is needed as this is specified by the customer. What we possibly miss in the training is the cooperation of the Universities with the LSPs to boost the practical aspect of machine translation post-editing and on one hand feed it with the quality aspects from the academic perspective and on the other hand balance it with the practical aspects of productivity, technology, and real life from the industrial perspective. The cooperation of LSPs and Universities with internships or other collaborative projects will also bring to front the role of the expert, the post-editor in this case, in the workflows and the importance of being able to handle the technology available.

The ISO Standard 18587 on MTPE

LG: "The post-editing standard (ISO 18587) has just been called for a systematic review [19.4.2022], and I think it’s perfect timing for that. It was published in 2017, and Google Translate had launched their neural machine translation only a few months before. So, things have changed a lot since then. There is a lot of talk going on about whether we should still talk about post-editing or not, if there should be not a review but a revamp of the standard of if we need it at all.

Viveta and I are very familiar with this and other standards (I am an ISO expert member representing my company and my country) and regardless of what happens to this ISO, the truth is that it is there for now and many companies have been certified with it, so it has inspired of course some of our discussions. For example, when we defined the two profiles of the post-editor (expert and junior) at the Special Interest Group, we considered the requirements from this standard. We also added an extra session to discuss how to define what post-editing is from a more academic point of view (because since in the Special Interest Group language service providers are, naturally, more represented than other groups, we had come up with a more industry-oriented definition)."

VG: "In the last calls of the Machine Translation Post-editing Training Special Interest Group, we discuss on the 5Ws and 1H, Who, What, Where, Why, When and How of machine translation post-editing, trying to reshape and redefine the service and have an agreed definition of the basic elements of the service from academic and industrial perspective. Based on the input and the voices heard, it seems that there is great room for the machine translation post-editing to be reviewed in the ISO as a service and that there is a long way to go ahead."

Hot topic: Remuneration for post-editors

LG: "Yes, it is a controversial topic, but that is why we thought we had to talk about it at the SIG. In Spanish we have an idiom which says ‘coger el toro por los cuernos’, which literally means ‘to take the bull by the horns’, that is to be brave and take control of a challenging situation.

That is what we tried to do with post-editing compensation, and we invested one full session to the topic. We presented first some examples of real comments from post-editors found on social media, we discussed about the aspects which affect the MTPE pricing (such as language pair and quality of the MT output), and we also presented the compensation options currently available in the market, such as pay per hour, word, based on effort (e. g. on edit distance), combination with TM, etc.

It is important to note that in our Special Interest Group we didn’t promote specifically any of these options, but instead we highlighted the pros & cons of each, and in the discussion per groups there was one thing that we all agreed with: any compensation method chosen needs to be transparent and supported by data."

VG: "A code of ethics for MTPE is needed and this is the topic of the last call in the Machine Translation Post-editing Training Special Interest Group."

The MTPE Training Protocol - Soon to be published

VG: "We will have the training protocol ready by the end of September 2022. In the Special Interest Group calls of April, May and June, we will touch the quintessence of the post-editing. These are the calls which will shape the training protocol for MTPE and we invite all members to join and be part of this vision with their contributions and ideas."

Sign up here for our newsletter on globalization and localization matters.