While still in high school, Xinyi Liu worked briefly in a lab at Beihang University in Beijing and was surprised to see Chinese researchers routinely using Google Translate to generate first draft scientific papers in English. Translation is mandatory if scientists want to submit high-profile journals, almost all of which are in English.
“It used to be normal for postdocs to just use Google Translate to first translate everything and then modify and polish it. But after the first translation, the whole paper doesn’t make sense,” said Liu, a rising junior at the University of California . , Berkeley, who is majoring in molecular and cell biology. “Literally, all the words, all the terms were just stuck together randomly.”
There had to be a better way, she thought.
So last year, when she saw a new seminar being taught by Rebecca Tarvin on breaking language barriers in science, she signed up.
That class, which will be taught at UC Berkeley for the third time in the spring of 2023, was a trial balloon for Tarvin, an assistant professor of integrative biology. With renewed campus-wide interest in diversity, equity and inclusion, she and working groups within her department thought the class could help UC Berkeley address a long-standing issue exists in science: English, the dominant language of science, is a major obstacle for scientists who are not native speakers of English.
It is not only foreign students and scientists who are at a disadvantage when science is communicated primarily in English. So are many American-born students. In fall 2020, approximately 40% of UC Berkeley’s first-generation students were first-generation college students, and within the 10-campus University of California system, 39% of first generation students grew up with a language other than English as their first language.
“Many of our students from California grew up translating for their parents,” Tarvin said. “Translation has been part of their lives since they were very young.”
For Tarvin, the class — Breaking Language Barriers in Evolution and Ecology — was “an opportunity both to teach students literacy skills in translation, and to encourage students to be activists in this area of structural change. And in fact, we’ve seen a really positive reception of this kind of activism from students, as they all seem to agree that addressing language barriers is really important after take the course.”
The class led Tarvin and some graduate students at UC Berkeley, along with collaborators in Canada, Israel and Hungary, to write a scientific paper evaluating new machine translation tools that can be used by people around the world. -world to make their scientific articles accessible to non-English speakers. . The paper appeared online this month in the journal BioScience. Translations into Spanish, French, Portuguese and Hungarian, the languages of the co-authors, are also online.
“The idea here is that we’re trying to give people the tools and the motivation to translate their own scientific research,” Tarvin said. “Science does not have to be based on one language. And there are many additional benefits that come from incorporating multilingual approaches in every phase of science. For example, publishing in several languages will be beneficial for society due to better communication of science.”
“Language can be a barrier, as well as a wonderful tool, to bring people together,” emphasized Emma Steigerwald, who is the first author of the paper and a UC Berkeley graduate student in science, politics and – environmental management. “It is an obstacle that we can overcome using this new technology. We explain about the technology and how it can be implemented and the things that we need to be aware of when using the technology, and the wonderful and positive ways all how science communication can be transformed by implementing this new technology”.
Towards a multilingual scientific network
Until recently, computer translation was the butt of jokes. People shared amusing examples of mistranslations, often seeming to disdain languages other than English and, by implication, other cultures.
But machine learning, or artificial intelligence, has dramatically increased the accuracy of translation to the point where tourists use Internet services like Google Translate to communicate with people in the countries they visit. .
But for a text that contains a lot of jargon — much of it scientific, but from many other academic fields, too — Google Translate is woefully inadequate.
“The quality of the translation is not for a journal,” said Ixchel Gonzalez Ramirez, one of the graduate student advisors for the course. “Many times, people have to pay to get a professional translator to translate their work, and that’s very expensive.”
The new paper highlights some of the many services — many of them free — that can convert English scientific writing into other languages. In addition to the well-known Google Translate platform, these include DeepL, which uses neural networks and claims to be many times more accurate than competitors when translating English into Chinese, Japanese, Romance languages or German, and vice versa; Baidu Translate, a service from the Chinese Internet company Baidu that initially focused on translation between English and Chinese; Naver Pagago, a multilingual translator created by a company in South Korea; and Yandex.Translate, which uses statistical machine translation and focuses mostly on Russian and English.
“Translation is becoming more and more accessible to every person. Whether you are an expert or not, and whether you are even bilingual or not, the ability to translate is so much easier with many of the technologies we have available today,” Steigerwald he said. “And so how can we integrate this into our workflow as scientists, and how does this change the expectations that surround scientific communication?”
English is the lingua franca of science
Tarvin’s interest in translation was sparked by one of her graduate students, Valeria Ramírez Castañeda, who in 2020 published a paper describing the costs incurred by her fellow Colombian doctoral students who wanted to publish or interact with ‘ colleagues in a world dominated by English.
As an evolutionary biologist interested in how some animals came to use venom, Tarvin decided to focus her new seminar on translating papers in the fields of evolution and ecology, although the students who signed up eventually they designed their own courses. She particularly looked for students, like Liu, and mentors, like Gonzalez Ramirez, who are bilingual or multilingual.
“Everyone in the class had some kind of relationship with the language related to the family,” said Tarvin.
Tarvin also asked Mairi-Louise McLaughlin, UC Berkeley professor of French and linguistics and an expert in journalistic and literary translation, to speak to the class about how professionals approach translation and how translation affects the meaning. That topic resonated with the students when they tried their hand at translating scientific abstracts and sometimes entire papers.
Ruoming Cui, a rising sophomore who took the course in the spring of 2022, chose Baidu to translate scientific abstracts. She immediately discovered that the long and complex sentences of English and the use of multiple words to describe a concept seemed redundant when delivered in Chinese.
“We don’t usually do that in Chinese because it will make every sentence extra long, and it’s very annoying,” she said.
Liu added that without considerable polishing, many English translations would be confused, she said.
“I’ve heard the saying that even though your result is amazing, if you write a confusing paper because of the translation, people will get annoyed because they can’t understand what you’re doing,” Liu said. “And that’s going to affect a lot how people validate the research or whether they’re even going to read it. I think that’s a big hurdle in the scientific world.”
Steigerwald, Tarvin and their co-authors also realized that writing scientific papers in plainer English — something that non-scientists have long encouraged — would equally benefit English speakers. -English and non-English speakers.
“If your first language is not English, and you are trying to read the English language version of the paper, it will feel much less ambiguous and much more readable when the writer has used simple language,” said Steigerwald. “But also, very importantly, when you go to translate that piece of text, machine learning tools are going to have a much easier time translating something that’s written in plain language. So, this is kind of kind of future-proof your writing, so that if someone wants to translate it into a million languages, they’ll have a much easier time of it when it’s written that way.”
Barriers to widespread translation of scientific articles remain, including where they are made available and how copyright is handled. Most journals do not even accept articles that are not in English, and few explicitly allow co-publishing of articles in translation. Tarvin found that few journals have any translation policies, and as a result of general copyright restrictions, many publishers charge exorbitant fees to post a translation online after publication.
“It’s quite amazing how many journals do not allow you to freely publish translations after publication, and how few have platform support where you can have even a simple abstract in a second or third language,” said Tarvin. “I think a major obstacle to this is the web platforms; not only the rules of publishing and copyright, but also the functionality of the platform.”
With the Breaking Barriers seminar and now the BioScience paper, Tarvin and her colleagues hope to gradually change the norm in science to not translate papers into other languages, especially the language of the country where the research was done and the languages of the co-authors.
And the more translations there are, the more material there is for training machine translation systems to do a better job, and the quality of scientific translation gradually increases.
“In my lab, we’re translating a lot of our research, and now the people in Emma’s lab are doing it too,” she said. “I think sharing our positive attitude towards this and how it can make a difference to people has influenced a small, but growing, group of people who are starting to incorporate translation into their scientific workflow. “
Additional co-authors of the BioScience paper include doctoral students Valeria Ramírez-Castañeda and Débora Brandt of UC Berkeley; András Báldi of the Institute of Ecology and Botany at the Center for Ecological Research in Vákrátót, Hungary; postdoctoral fellow Julie Teresa Shapiro of Ben-Gurion University of the Negev in Be’er Sheva, Israel; and Lynne Bowker, professor of translation and interpreting at the University of Ottawa in Canada.
Which field of AI is used for translation?
It’s Called Machine Translation AI-assisted translation is called machine translation. Read also : The Alabama State Council on the Arts provides fellowship to four UAB trainers and staff. What it does is convert text from the source language to text in the output language.
What area of AI did the company use to develop Google translate? It is called Google Neural Machine Translation system (GNMT) and basically, it means that the system can learn from what is fed into it.
Which type of artificial intelligence used for translating one language to another?
Neural machine translation (NMT)
What benefits do you see of using a machine learning AI translator rather than a human translator?
The advantages of machine translation On the same subject : Professor: Music brings children out best.
- Many free tools are readily available (Google Translate, Skype Translator, etc.)
- Quick turnaround time.
- You can translate between multiple languages using one tool.
- Translation technology is constantly improving.
What is the difference between human translation and machine translation? Machine translation â background, evolution, and types Well, machine translation is the immediate transformation of text from one language to another using artificial intelligence. Human translation, on the other hand, involves real brainpower, in the form of one or more translators translating the text manually.
Why is machine translation better than human?
Machine translation tends to ignore quality because it cannot relate and understand the format of the writing. The human translation ensures that the quality is achieved, by recreating the document in a similar format. On the same subject : Prairie Heights offers Civic Arts graduation route | Heraldrepublican. The machine translator always confuses a simple word, making it difficult to read and understand.
Will machine translation get better?
Machine Translation can increase the capabilities of translators by 3-5 times in some cases, allowing more content to be localized in a shorter amount of time.
What is the future of machine translation? Machine Translation can increase the capabilities of translators by 3-5 times in some cases, allowing more content to be localized in a shorter amount of time. With increased productivity and reduced costs, companies will be able to translate more content into more languages.
Is machine translation getting better?
Machine translators are becoming more efficient and the quality of the translations produced is becoming more understandable. However, the need for humans to be part of the machine translation equation is always there.
Is there a future in translation?
Unlike many other sectors, translation has weathered the recession with ease, and continues to expand. A recent report from the CSA (Common Sense Advisory, ‘Top 100 Language Service Providers: 2016’) estimates that the translation industry will have grown to USD 45 billion by 2020.
How are machine translation systems implemented?
The basic approach involves connecting the input sentence structure to the output sentence structure using a parser and analyzer for the source language, a generator for the target language, and a lexicon of ‘ transfer to the current translation.
What are the main stages of machine translation? Therefore, SMT requires three steps: 1) Language Model (what is the right word given its context?); 2) Translation Model (what is the best translation of a given word?); 3) a method to find the correct word order. In the previous paragraphs, we used both the sentence and the word as the unit of translation.
How does a translation machine work?
How Machine Translation Works. Machine translation (MT) is automated, meaning it is the translation of the text by a computer without any human involvement. It works by using computer software to translate text from one language (source language) to another language (target language).
