machine translation
In plain wordsSoftware that renders writing or speech in one language into another language, as online translation services and phone translation apps do.
The automatic translation of text or speech between natural languages. Approaches have passed through rule-based systems (bilingual dictionaries plus hand-written transfer rules or an interlingua), statistical machine translation (from the late 1980s, learning word- and phrase-alignment probabilities from parallel corporacorpus, n.: A large collection of written or spoken language, gathered so that people or computers can study it or learn from it — like a library assembled for a language model.), neural machine translation (from 2014, end-to-end encoder–decoderencoder–decoder architecture, n.: A two-part model design: one part, the encoder, reads the input and condenses it into an internal summary; the other, the decoder, uses that summary to produce the output. networks with attentionattention, n.: A way for an AI model to choose what to focus on. For each word it processes, the model scores how relevant every other word is and draws information mostly from the relevant ones.) and, since the early 2020s, large language modelslarge language model, n.: A computer program trained on enormous amounts of text to predict the words that come next. At vast scale, that simple skill lets it chat, answer questions, translate, summarize and write code. that translate as one capability among many. Quality is assessed by human raters and by automatic metrics such as BLEUBLEU, abbr.: An automatic score for machine translation that checks how many short runs of words in the computer's translation also appear in human reference translations. and newer learned metrics. Translation was a founding problem for NLP, and the transformertransformer, n.: The design behind today's chatbots and much of modern AI. It handles text by letting every word look at every other word and decide which of them matter for understanding it. was introduced in a translation paper. Remaining challenges include low-resource languageslow-resource language, n.: A language with little digital text, recorded speech or language tools available — often one spoken by millions — which makes AI systems much weaker in it than in English., document-level consistency, idiom and terminology, and biases such as defaulting to gendered forms.
Machine translation handles the first draft; human linguists post-edit the legal clauses.
Origin
Proposed in Warren Weaver's 1949 memorandum 'Translation'; the Georgetown–IBM experiment of 7 January 1954 translated more than 60 Russian sentences on an IBM 701. The 1966 ALPAC reportALPAC report, prop. n.: A 1966 US government report that concluded machine translation was slower, worse and costlier than human translators. It ended most American funding for the field for many years. cut US funding. IBM's statistical models (late 1980s–1993) and neural sequence-to-sequence models (2014) set the later course, and Google Translate moved to neural MT in 2016.
Related
- natural language processingnatural language processing, n.: The field that teaches computers to handle human language — to read, interpret, translate, summarize and produce text and speech. It is one of the oldest and largest areas of artificial intelligence.
- BLEUBLEU, abbr.: An automatic score for machine translation that checks how many short runs of words in the computer's translation also appear in human reference translations.
- sequence-to-sequence modelsequence-to-sequence model, n.: A model that turns one sequence into another, such as a French sentence into English or speech into text, by first reading the whole input and then writing the output step by step.
- attentionattention, n.: A way for an AI model to choose what to focus on. For each word it processes, the model scores how relevant every other word is and draws information mostly from the relevant ones.
- Attention Is All You NeedAttention Is All You Need, prop. n.: The 2017 Google research paper that introduced the transformer, the neural-network design behind ChatGPT, Claude, Gemini and nearly every other modern AI language system.
- ALPAC reportALPAC report, prop. n.: A 1966 US government report that concluded machine translation was slower, worse and costlier than human translators. It ended most American funding for the field for many years.
- low-resource languagelow-resource language, n.: A language with little digital text, recorded speech or language tools available — often one spoken by millions — which makes AI systems much weaker in it than in English.
Sources
- Brown, P. F., Della Pietra, S. A., Della Pietra, V. J. & Mercer, R. L. (1993). The Mathematics of Statistical Machine Translation: Parameter Estimation. Computational Linguistics 19(2), 263–311. aclanthology.org/J93-2003
- Bahdanau, D., Cho, K. & Bengio, Y. (2015). Neural Machine Translation by Jointly Learning to Align and Translate. ICLR 2015.
- Papineni, K., Roukos, S., Ward, T. & Zhu, W.-J. (2002). Bleu: a Method for Automatic Evaluation of Machine Translation. ACL 2002. aclanthology.org/P02-1040
Mentioned in
natural language processingnatural language processing, n.: The field that teaches computers to handle human language — to read, interpret, translate, summarize and produce text and speech. It is one of the oldest and largest areas of artificial intelligence., Association for Computational LinguisticsAssociation for Computational Linguistics, prop. n.: The scientific society at the center of research on language technology, from translation software to chatbots. It runs the field's leading conferences and a free online archive of their papers., Attention Is All You NeedAttention Is All You Need, prop. n.: The 2017 Google research paper that introduced the transformer, the neural-network design behind ChatGPT, Claude, Gemini and nearly every other modern AI language system., BLEUBLEU, abbr.: An automatic score for machine translation that checks how many short runs of words in the computer's translation also appear in human reference translations., computational linguisticscomputational linguistics, n.: The science of human language studied with computers: using computational models to understand how language works, and building the language technology that grows out of that understanding., end-to-end learningend-to-end learning, n.: Training one AI system to go straight from raw input, such as camera images, to the final output, such as steering commands, instead of chaining hand-designed steps., Ilya SutskeverIlya Sutskever, prop. n.: A Russian-born Israeli-Canadian computer scientist who co-built the AlexNet image recognizer, helped found OpenAI and lead its research until 2024, and now heads the startup Safe Superintelligence Inc., long short-term memorylong short-term memory, n.: A type of recurrent network with a separate memory lane and learned 'gates' that decide what to remember, forget and output, letting it keep track of information over long sequences., low-resource languagelow-resource language, n.: A language with little digital text, recorded speech or language tools available — often one spoken by millions — which makes AI systems much weaker in it than in English., natural language generationnatural language generation, n.: Getting computers to produce human language — from turning a spreadsheet into a weather report to writing an essay or a chatbot's reply., sequence-to-sequence modelsequence-to-sequence model, n.: A model that turns one sequence into another, such as a French sentence into English or speech into text, by first reading the whole input and then writing the output step by step., ALPAC reportALPAC report, prop. n.: A 1966 US government report that concluded machine translation was slower, worse and costlier than human translators. It ended most American funding for the field for many years., Ashish VaswaniAshish Vaswani, prop. n.: An Indian-born computer scientist listed first on 'Attention Is All You Need', the 2017 paper that introduced the transformer, the design behind today's language models., Dzmitry BahdanauDzmitry Bahdanau, prop. n.: A Belarusian computer scientist who, as a student in Yoshua Bengio's lab in 2014, was lead author of the paper that introduced attention, the mechanism later at the core of the transformer., Frederick JelinekFrederick Jelinek, prop. n.: A Czech-born American engineer who led IBM's speech-recognition research and made statistics learned from data, rather than hand-written grammar rules, the basis of language technology., Georgetown–IBM experimentGeorgetown–IBM experiment, prop. n.: A 1954 public demonstration in New York in which an IBM computer translated about sixty Russian sentences into English. It raised hopes for automatic translation that took decades to fulfill., IBM ResearchIBM Research, prop. n.: The research division of the computer company IBM, behind landmark AI systems such as the chess champion Deep Blue and the quiz-show winner Watson., Kyunghyun ChoKyunghyun Cho, prop. n.: A South Korean computer scientist who co-invented the encoder–decoder approach to neural machine translation and the gated recurrent unit, and co-wrote the paper that introduced attention., word sense disambiguationword sense disambiguation, n.: Working out which meaning of a word is intended in context — whether 'bank' means a riverside or a place that keeps money.