Emily M. Bender | |
---|---|
Born | 1973 (age 50–51) |
Known for | Research on the risks of large language models and ethics of NLP; coining the term 'Stochastic parrot'; research on the use of Head-driven phrase structure grammar in computational linguistics |
Spouse | Vijay Menon [1] |
Parent |
|
Academic background | |
Alma mater | UC Berkeley and Stanford University [3] [4] |
Thesis | Syntactic variation and linguistic competence: The case of AAVE copula absence (2000 [3] [4] ) |
Doctoral advisor | Tom Wasow Penelope Eckert [4] |
Academic work | |
Discipline | Linguistics |
Sub-discipline | Syntax,computational linguistics |
Institutions | University of Washington |
Emily Menon Bender (born 1973) is an American linguist who is a professor at the University of Washington. She specializes in computational linguistics and natural language processing. She is also the director of the University of Washington's Computational Linguistics Laboratory. [5] [6] She has published several papers on the risks of large language models and on ethics in natural language processing. [7]
Bender earned an AB in Linguistics from UC Berkeley in 1995. She received her MA from Stanford University in 1997 and her PhD from Stanford in 2000 for her research on syntactic variation and linguistic competence in African American Vernacular English (AAVE). [8] [3] She was supervised by Tom Wasow and Penelope Eckert. [4]
Before working at University of Washington,Bender held positions at Stanford University,UC Berkeley and worked in industry at YY Technologies. [9] She currently holds several positions at the University of Washington,where she has been faculty since 2003,including professor in the Department of Linguistics,adjunct professor in the Department of Computer Science and Engineering,faculty director of the Master of Science in Computational Linguistics, [10] and director of the Computational Linguistics Laboratory. [11] Bender is the current holder of the Howard and Frances Nostrand Endowed Professorship. [12] [13]
Bender was elected VP-elect of the Association for Computational Linguistics in 2021. [14] Bender served as VP-elect in 2022,moving to Vice-President in 2023. She is serving as President through 2024, [15] [16] and will serve as Past President in 2025. Bender was elected a Fellow of the American Association for the Advancement of Science in 2022. [17]
Bender has published research papers on the linguistic structures of Japanese,Chintang,Mandarin,Wambaya,American Sign Language and English. [18]
Bender has constructed the LinGO Grammar Matrix,an open-source starter kit for the development of broad-coverage precision HPSG grammars. [19] [20] In 2013,she published Linguistic Fundamentals for Natural Language Processing:100 Essentials from Morphology and Syntax, and in 2019,she published Linguistic Fundamentals for Natural Language Processing II:100 Essentials from Semantics and Pragmatics with Alex Lascarides,which both explain basic linguistic principles in a way that makes them accessible to NLP practitioners.[ citation needed ]
In 2021,Bender presented a paper,"On the Dangers of Stochastic Parrots:Can Language Models Be Too Big? 🦜" co-authored with Google researcher Timnit Gebru and others at the ACM Conference on Fairness,Accountability,and Transparency [21] that Google tried to block from publication,part of a sequence of events leading to Gebru departing from Google,the details of which are disputed. [22] The paper concerned ethical issues in building natural language processing systems using machine learning from large text corpora. [23] Since then,she has invested efforts to popularize AI ethics and has taken a stand against hype over large language models. [24] [25]
The Bender Rule,which originated from the question Bender repeatedly asked at the research talks,is research advice for computational scholars to "always name the language you're working with". [1]
She draws a distinction between linguistic form versus linguistic meaning. [1] Form refers to the structure of language (e.g. syntax),whereas meaning refers to the ideas that language represents. In a 2020 paper,she argued that machine learning models for natural language processing which are trained only on form,without connection to meaning,cannot meaningfully understand language. [26] Therefore,she has argued that tools like ChatGPT have no way to meaningfully understand the text that they process,nor the text that they generate.[ citation needed ]
Computational linguistics is an interdisciplinary field concerned with the computational modelling of natural language, as well as the study of appropriate computational approaches to linguistic questions. In general, computational linguistics draws upon linguistics, computer science, artificial intelligence, mathematics, logic, philosophy, cognitive science, cognitive psychology, psycholinguistics, anthropology and neuroscience, among others.
Natural language processing (NLP) is an interdisciplinary subfield of computer science and information retrieval. It is primarily concerned with giving computers the ability to support and manipulate human language. It involves processing natural language datasets, such as text corpora or speech corpora, using either rule-based or probabilistic machine learning approaches. The goal is a computer capable of "understanding" the contents of documents, including the contextual nuances of the language within them. To this end, natural language processing often borrows ideas from theoretical linguistics. The technology can then accurately extract information and insights contained in the documents as well as categorize and organize the documents themselves.
In linguistics, syntax is the study of how words and morphemes combine to form larger units such as phrases and sentences. Central concerns of syntax include word order, grammatical relations, hierarchical sentence structure (constituency), agreement, the nature of crosslinguistic variation, and the relationship between form and meaning (semantics). There are numerous approaches to syntax that differ in their central assumptions and goals.
Head-driven phrase structure grammar (HPSG) is a highly lexicalized, constraint-based grammar developed by Carl Pollard and Ivan Sag. It is a type of phrase structure grammar, as opposed to a dependency grammar, and it is the immediate successor to generalized phrase structure grammar. HPSG draws from other fields such as computer science and uses Ferdinand de Saussure's notion of the sign. It uses a uniform formalism and is organized in a modular way which makes it attractive for natural language processing.
Parsing, syntax analysis, or syntactic analysis is the process of analyzing a string of symbols, either in natural language, computer languages or data structures, conforming to the rules of a formal grammar. The term parsing comes from Latin pars (orationis), meaning part.
Ivan Andrew Sag was an American linguist and cognitive scientist. He did research in areas of syntax and semantics as well as work in computational linguistics.
John Robert "Haj" Ross is an American poet and linguist. He played a part in the development of generative semantics along with George Lakoff, James D. McCawley, and Paul Postal. He was a professor of linguistics at MIT from 1966 to 1985 and has worked in Brazil, Singapore and British Columbia, and until spring 2021, he taught at the University of North Texas.
Joan Wanda Bresnan FBA is Sadie Dernham Patek Professor in Humanities Emerita at Stanford University. She is best known as one of the architects of the theoretical framework of lexical functional grammar.
Eva Hajičová [ˈɛva ˈɦajɪt͡ʃovaː] is a Czech linguist, specializing in topic–focus articulation and corpus linguistics. In 2006, she was awarded the Association for Computational Linguistics (ACL) Lifetime Achievement Award. She was named a fellow of the ACL in 2011.
Google Brain was a deep learning artificial intelligence research team under the umbrella of Google AI, a research division at Google dedicated to artificial intelligence. Formed in 2011, it combined open-ended machine learning research with information systems and large-scale computing resources. It created tools such as TensorFlow, which allow neural networks to be used by the public, and multiple internal AI research projects, and aimed to create research opportunities in machine learning and natural language processing. It was merged into former Google sister company DeepMind to form Google DeepMind in April 2023.
Georgia M. Green is an American linguist and academic. She is an emeritus professor at the University of Illinois at Urbana-Champaign. Her research has focused on pragmatics, speaker intention, word order and meaning. She has been an advisory editor for several linguistics journals or publishers and she serves on the usage committee for the American Heritage Dictionary.
Ellen M. Kaisse is an American linguist. She is professor emerita of linguistics at the University of Washington, best known for her research on the interface between phonology, syntax, and morphology.
Raffaella Zanuttini is an Italian linguist whose research focuses primarily on syntax and linguistic variation. She is a Professor of Linguistics at Yale University in New Haven, Connecticut.
Mirella Lapata FRSE is a computer scientist and Professor in the School of Informatics at the University of Edinburgh. Working on the general problem of extracting semantic information from large bodies of text, Lapata develops computer algorithms and models in the field of natural language processing (NLP).
Timnit Gebru is an Eritrean Ethiopian-born computer scientist who works in the fields of artificial intelligence (AI), algorithmic bias and data mining. She is an advocate for diversity in technology and co-founder of Black in AI, a community of Black researchers working in AI. She is the founder of the Distributed Artificial Intelligence Research Institute (DAIR).
The usage-based linguistics is a linguistics approach within a broader functional/cognitive framework, that emerged since the late 1980s, and that assumes a profound relation between linguistic structure and usage. It challenges the dominant focus, in 20th century linguistics, on considering language as an isolated system removed from its use in human interaction and human cognition. Rather, usage-based models posit that linguistic information is expressed via context-sensitive mental processing and mental representations, which have the cognitive ability to succinctly account for the complexity of actual language use at all levels. Broadly speaking, a usage-based model of language accounts for language acquisition and processing, synchronic and diachronic patterns, and both low-level and high-level structure in language, by looking at actual language use.
Margaret Mitchell is a computer scientist who works on algorithmic bias and fairness in machine learning. She is most well known for her work on automatically removing undesired biases concerning demographic groups from machine learning models, as well as more transparent reporting of their intended use.
Mona Talat Diab is a computer science professor and director of Carnegie Mellon University's Language Technologies Institute. Previously, she was a professor at George Washington University and a research scientist with Facebook AI. Her research focuses on natural language processing, computational linguistics, cross lingual/multilingual processing, computational socio-pragmatics, Arabic language processing, and applied machine learning.
Vera Demberg is a German computational linguist and professor of computer science and computational linguistics at Saarland University.
In machine learning, the term stochastic parrot is a metaphor to describe the theory that large language models, though able to generate plausible language, do not understand the meaning of the language they process. The term was coined by Emily M. Bender in the 2021 artificial intelligence research paper "On the Dangers of Stochastic Parrots: Can Language Models Be Too Big? 🦜" by Bender, Timnit Gebru, Angelina McMillan-Major, and Margaret Mitchell.