How will the development of chat GPT and LLM help the development of machine translation
Getting started
If you're a translator, I think you'll be very much appreciated in the Chat GPT and LLM I explained in today's title. In fact, Chat GPT is also a topic you might be surprised in even if you're not working in the translation industry. It's also a part that has an impact on a wide range of industries. Since machine translation attempt is an AI-based task, the development of Chat GPT is also related to the translation god. So today, I'm going to write about how the development of such technology can help with translation.
What is an LLM
LLM is an abbreviation of large language model and can be understood as a super-large language model, etc. The expected impact of LLM on machine translation is that it is a deep learning algorithm that can perform various natural language processing tasks. It's a deep learning model that can understand and generate text through training. Through this learning, you can answer questions based on knowledge you know, start with text classification, and even tasks such as copywriting. So how can LLM impact machine translation performance?
What is Chat GPT
Chat GPT is a large language model, or LLM, for natural language processing. The use of chat GPT can be used in natural language processing tasks such as chatbots, classification, summary reporting, and question and answer systems that intersect answer any question arguably of field, and is actually being used. Everything discussed about LLM in the previous paragraph is possible with Chat GPT. In fact, many users use chat GPT in real life, work, or study. This technology, which explains LLM, provides a path forward for machine translation. It is now reduced to go through post-editing work after machine translation, but I think that if this technology is developed, the time costs reduced in post-processing will be.
The correlation between the development of LLM and translation
Everyone knows that advances in both technologies can help with translation. If so, where exactly can we help? I was fascinated by the fact that LLM Technology “Imperfect Imperfect Fears” is more than any other function. This suggests like an essential feature that could help with translation. Not only when writing, but also when speaking, there are times when I don't write or utter words that are fully equipped with sentence elements.
Just like when writing a novel, if the subject is omitted in the previous sentence, it may be omitted in the next sentence. The same goes for speaking. If you know what or who the conversation people are talking about, that part can be omitted. At this point, machine translation problems are often discussed. Each segment is translated, but if sentence elements are missing, they are sometimes filled in and translated. I think LLM can reduce errors that happened at this time. Can you predict the subject in the next segment by considering to the subject in the previous sentence. Because of this aspect, I'm hopeful that more than anything else, I can help with video translation. I was hopeful that it would reduce the energy consumption on company-wide work. I will finish today's notes looking forward to LLM, which will be more advanced than it is now, and machine translation, which will be more complete along with it.
Editor/Lee In Hye
Inhye.lee@twigfarm.net
