Terminology-aware machine translation in Trados 2024

Question

Dear Community! 
 We are currently analyzing the upgrade of current 2022 licenses to 2024, therefore studying all the novelties offered by this new version. 
 According to the information available, MT results can be modified by the terminology available in the Multiterm dictionary, isn&acute;t it? 
 So as far as I can understand, the working principle is the same as with the DeepL glossaries (which I personally prefer to call "MT correction lists", cause they are 100% personalized to the type of text translated and provide almost no information for a better understanding of a specific term). Or are there other benefits that I don&acute;t see? 
 However, the "handicap" I see is how the MT provider will "choose" the term from the list of terms available in the Multiterm database? 
 Practical example 1: 
 PLANO (Spanish term): in our corporate MDB we have three entries for "plano": 1) wheel tread defect 2) plane (vertical, horizontal...) and 3 ) drawing. 
 In this particular case, which term would the AI Assistant 2024 be aware???? 
 Practical case 2: 
 Bocina de agudos: this entry has 3 synonym terms in English: high-pitch horn, high-pitched horn and whistle. 
 Which synonym of the three would the AI Assistant use in this case? 
 DeepL glossaries are more specific, i.e. they allow to correct a specific word with another specific word, only one. 
 However, if the Multiterm glossary is applied in Trados 2024 following the same principle...how will the algorythm work? And then, can Multiterm termonology be used on the MT provided by DeepL? Or is is mandatory to use other MT resources (Azure, Open Ai, etc?) 
 Thanks in advance!

Daniel Hug · Answer

Yulia Kovalskaya 
 a) Yes, that is the AI Assistant in Trados 2024. 
 b) The way I set it up, you have to click to insert it. But you could automate this and use it as a batch task. As Paul pointed out, the current AIs are fuzzy &ndash; that's their glory and their weakness. So you will want to look at what it's doing, at the very least check the results. 
 c) One prompt, hastily written in this case. It makes sense to refine prompts. I could imagine customizing a prompt for a project, like &ldquo;The text you are working on is about cleaning textiles.&rdquo; If you can give the LLM relevant information, by all means do. 
 
 The one serious weakness I see is how forbidden terms are handled. They, too, are just passed on to AI as relevant terms. If your termbase has a lot of those, you might have to create a separate termbase to get rid of those. In the above example, if I had "Klecks" as forbidden term, I doubt AI could reliably select "Fleck". All it get is a list like: "stain-Fleck, stain-Klecks, stain-Beize". Across the board, handling forbidden terms is a topic that urgently needs attention IMHO. 
 But as for your main question, AI's terminology-aware translations are seriously above a DeepL-style glossary if you prompt it well. As I said in other posts before, I believe this is a big game-changer technology for this industry, just like the ascent of MT was ten years ago. 
 Would be worth checking whether a plain terminology-redacting task requires GPT-4o, or whether it can run on a simpler, faster and cheaper model like 3.5turbo. There is so much that could be done! 
 Daniel

Daniel Hug · Answer

Yulia Kovalskaya 
 Yulia Kovalskaya said: the working principle is the same as with the DeepL glossaries 
 Well, "jein", as we might say in German. Yes and no. At this point, the AI is just given the recognized terms, without any metadata. Insofar yes, your termbase is in a way reduced to a glossary. I hope this will change and there will be a way to choose other fields from the termbase to be handed to the AI. 
 If you prompt the AI to give you several suggestions, you will see that it varies which of the synonyms it uses. And you can instruct it to use a fitting synonym. Insofar, no, it's not like a DeepL glossary. That will always slot in the first fitting term it comes across, even if it makes little sense. With the AI, you can instruct it how to select a synonym. You might have to custom tailor your prompt for a particular translation to tell AI what the context is, but even a general instruction works. See screenshots below, I created a termbase with two translations for the English word "stain", and gave it two sentences. I also instructed AI to explain its choice. 
 In the first sentence, MT got the correct term, which is accepted by AI. In the second sentence, MT got the wrong term, and AI corrects it:

This is just a quick test. I am sure this could be refined substantially.

Paul · Answer

Yulia Kovalskaya 
 Yulia Kovalskaya said: So as far as I can understand, the working principle is the same as with the DeepL glossaries (which I personally prefer to call "MT correction lists", cause they are 100% personalized to the type of text translated and provide almost no information for a better understanding of a specific term). Or are there other benefits that I don´t see? 
 Similar... but not the same. 
 DeepL changes the terms at the point of receiving the source so the MT results returned already contains the corrected term. 
 The AI solution corrects the MT results that has been returned based on the terminology you have in your termbase. But since it's AI you also benefit from the AI ensuring that the term is grammatically accurate in the context of the overall sentence. 
 Yulia Kovalskaya said: However, the "handicap" 
 That is a good question. I would typically respond by saying it's up to you to ensure you have a data structure that is suitable to support a solution like this. I'm not sure how well the AI would do in ensuring the correct term is used in your example... it could lead to some odd translations. Longer term we have discussed seeing how best to improve this but for now I think the solution to be sure (as sure as you can be with a tool that may not deliver consistent results 100% of the time due to the nature of AI right now) is to work on your termbase. 
 But definitely worth playing with it a bit to see what sort of results you get. 
 Yulia Kovalskaya said: And then, can Multiterm termonology be used on the MT provided by DeepL? 
 Yes it can. You can use it with any MT provider.

Trados Studio > 3. MultiTerm

Terminology-aware machine translation in Trados 2024

Top Replies