Skip to content
Last Dictionary
Plane
Language Models & TransformersPhilosophy & MindSymbolic & Classical AI

natural language understanding

nounAlso: NLU, language understanding, intent detection, slot filling

In plain wordsThe part of language AI concerned with working out what a piece of text means — who did what to whom, what the writer wants and what follows from it — rather than with producing text.

  1. 1

    The subfield of 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. concerned with mapping text, or transcribed speech, to representations of its meaning that support action or inference: identifying entities, relations and events, resolving references and ambiguity, inferring intent and entailments, and answering questions. Benchmarks such as GLUEGLUE, prop. n.: A 2018 bundle of nine English language-understanding tests combined into one score, which became the yardstick for the first wave of pretrained language models such as BERT. (General Language Understanding Evaluation, 2018), SuperGLUESuperGLUE, prop. n.: A harder follow-up to the GLUE language-understanding test, released in 2019 after AI systems had already beaten the human baseline on the original. and the Winograd schema challengeWinograd schema challenge, prop. n.: A test of machine common sense built from pairs of sentences in which changing one word changes what a pronoun refers to — something people resolve instantly using everyday knowledge. were built to measure it, and natural language inferencenatural language inference, n.: Deciding whether one sentence follows from another, contradicts it, or neither: if 'a man is playing a guitar on stage', is it true that 'a man is performing music'?, question answeringquestion answering, n.: Systems that reply to people's questions with the right answer, whether by finding it in a passage they are given, searching a database or the web, or drawing on what they already know. and coreference resolutioncoreference resolution, n.: Working out which words in a text are talking about the same someone or something — for instance, that 'she', 'the senator' and 'Ms. Okafor' all mean one woman. are its staple tests. In the classic division of NLP it is paired with 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.; encoder models such as BERTBERT, prop. n.: An influential 2018 Google language model that learned to understand text by filling in hidden words. It made search engines and many language tools much better at grasping meaning in context. were long described as NLU models and decoder models as generators, a distinction that large language models have largely dissolved.

    The benchmark is supposed to test natural language understanding, but models exploit annotation artifacts.

  2. 2

    Conversational systems

    In voice assistants and customer-service bots, the component that converts a user's utterance into a structured intent and slot values (an intent such as book_flight with destination set to Lisbon) for downstream logic — a narrower engineering sense still common in industry.

Origin

Established in AI by the early 1970s: Terry Winograd's 1971 MIT thesis, published as 'Understanding Natural Language' (1972), described the SHRDLUSHRDLU, prop. n.: A program from around 1970 that let people move toy blocks in a simulated world by typing ordinary English commands and questions, and that could explain what it had done and why. system, and Roger Schank's conceptual-dependency work pursued meaning representation in the same period. GLUE (2018) gave the term its benchmark-era sense.

Usage note

Whether statistical models genuinely understand language is a long-running dispute (see machine understandingmachine understanding, n.: Whether, and in what sense, an AI system grasps what words, situations or problems mean — rather than just producing answers that look as though it does., octopus testoctopus test, n.: A thought experiment in which a clever octopus learns to impersonate a stranded islander by eavesdropping on an undersea cable, without ever knowing what the words refer to. It asks whether text-trained AI can grasp meaning. and Chinese roomChinese room, n.: A thought experiment: a person who knows no Chinese follows a rulebook to answer Chinese messages so well that outsiders think they understand. If the person doesn't understand, does a computer running a program?); in technical usage NLU names a family of tasks, not a claim about comprehension.

Related

See also
  • 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.
  • 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.
  • natural language inferencenatural language inference, n.: Deciding whether one sentence follows from another, contradicts it, or neither: if 'a man is playing a guitar on stage', is it true that 'a man is performing music'?
  • GLUEGLUE, prop. n.: A 2018 bundle of nine English language-understanding tests combined into one score, which became the yardstick for the first wave of pretrained language models such as BERT.
  • machine understandingmachine understanding, n.: Whether, and in what sense, an AI system grasps what words, situations or problems mean — rather than just producing answers that look as though it does.
  • SHRDLUSHRDLU, prop. n.: A program from around 1970 that let people move toy blocks in a simulated world by typing ordinary English commands and questions, and that could explain what it had done and why.
  • question answeringquestion answering, n.: Systems that reply to people's questions with the right answer, whether by finding it in a passage they are given, searching a database or the web, or drawing on what they already know.
Contrast with
  • 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.

Sources

  1. Winograd, T. (1972). Understanding Natural Language. Academic Press.
  2. Wang, A., Singh, A., Michael, J., Hill, F., Levy, O. & Bowman, S. R. (2018). GLUE: A Multi-Task Benchmark and Analysis Platform for Natural Language Understanding. BlackboxNLP Workshop, EMNLP 2018.

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., conversational AIconversational AI, n.: The technology that lets computers hold conversations with people in everyday language, by text or voice. It powers chatbots, voice assistants and automated customer-service lines., GLUEGLUE, prop. n.: A 2018 bundle of nine English language-understanding tests combined into one score, which became the yardstick for the first wave of pretrained language models such as BERT., machine understandingmachine understanding, n.: Whether, and in what sense, an AI system grasps what words, situations or problems mean — rather than just producing answers that look as though it does., 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., spurious correlationspurious correlation, n.: A link in data that fools you: two things move together, yet neither drives the other, because a hidden shared cause, the way the data were collected or luck produced the pattern. AI models readily latch onto such links., virtual assistantvirtual assistant, n.: Software you speak or type to, like Siri, Alexa or Google Assistant, that answers questions and does simple jobs such as setting timers, playing music or controlling smart-home devices., coreference resolutioncoreference resolution, n.: Working out which words in a text are talking about the same someone or something — for instance, that 'she', 'the senator' and 'Ms. Okafor' all mean one woman., natural language inferencenatural language inference, n.: Deciding whether one sentence follows from another, contradicts it, or neither: if 'a man is playing a guitar on stage', is it true that 'a man is performing music'?, Percy LiangPercy Liang, prop. n.: An American computer scientist at Stanford whose center coined the term 'foundation model' and builds tools for evaluating and documenting large AI models., SHRDLUSHRDLU, prop. n.: A program from around 1970 that let people move toy blocks in a simulated world by typing ordinary English commands and questions, and that could explain what it had done and why., SuperGLUESuperGLUE, prop. n.: A harder follow-up to the GLUE language-understanding test, released in 2019 after AI systems had already beaten the human baseline on the original., Terry WinogradTerry Winograd, prop. n.: An American computer scientist who built SHRDLU, a celebrated early program that conversed in English about a world of blocks, and later became a critic of symbolic AI., 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.