Conference on Empirical Methods in Natural Language Processing

Conference on Empirical Methods in Natural Language Processing
Name	Conference on Empirical Methods in Natural Language Processing
Acronym	EMNLP
Established	1996
Discipline	Natural language processing
Publisher	Association for Computational Linguistics
Frequency	Annual

Contents

Conference on Empirical Methods in Natural Language Processing is an annual international academic conference focused on computational approaches to human language, benchmarking statistical and machine learning methods for textual and speech data. The conference draws researchers from institutions such as Massachusetts Institute of Technology, Stanford University, University of Oxford, Carnegie Mellon University and University of Cambridge, and is commonly associated with professional organizations like the Association for Computational Linguistics, International Speech Communication Association, and research labs including Google Research, Microsoft Research, Facebook AI Research and DeepMind. EMNLP has become a central venue alongside ACL (conference), NAACL, and COLING for presenting empirical advances in language technologies.

History

EMNLP originated in the mid-1990s amid a shift toward data-driven methods influenced by work at institutions such as IBM Research, Bell Labs, AT&T Laboratories, SRI International and universities including University of Pennsylvania and University of Edinburgh. Early predecessors and related gatherings included workshops linked to ACL (conference), IJCAI, and NeurIPS where researchers from University of California, Berkeley, University of Washington, University of Toronto, and Princeton University exchanged corpus-based studies. Over successive decades, EMNLP absorbed threads from speech and information retrieval communities associated with TREC and ICASSP, and adapted to paradigm shifts propelled by advances at OpenAI, Google DeepMind, Facebook AI Research, and major labs at Alibaba Group and Baidu Research. Key organizational actors have included the Association for Computational Linguistics and program chairs drawn from Johns Hopkins University, University of Illinois Urbana–Champaign, Tsinghua University, and Peking University.

EMNLP covers empirical and statistical methods for language, often featuring work related to machine learning developments from NeurIPS, ICML, and ICLR; language resources from LREC; and evaluation frameworks akin to those used at BLEU-driven evaluation meetings. Typical topics span neural architectures influenced by research at Google Brain and DeepMind, pretrained models following breakthroughs like BERT, GPT-3, and RoBERTa, and tasks including machine translation advanced by teams at Facebook AI Research and Microsoft Translator, question answering explored at Allen Institute for AI and Stanford Question Answering Dataset, information extraction informed by work at Columbia University, and dialog systems connected to projects at Amazon Alexa and Apple Siri. Cross-cutting themes include fairness investigations resonant with studies at Oxford Internet Institute and Harvard University, multilinguality linked to initiatives at United Nations Educational, Scientific and Cultural Organization-associated corpora, and evaluation methods paralleling efforts at NIST and DARPA.

EMNLP is organized under the auspices of professional bodies such as the Association for Computational Linguistics with steering committees composed of senior researchers from Stanford University, Carnegie Mellon University, University of Oxford, University of California, Berkeley, and ETH Zurich. Program committees often include scholars affiliated with University of Toronto, Imperial College London, University of Melbourne, National University of Singapore, and industry labs at Google Research, Microsoft Research, Facebook AI Research, and Alibaba DAMO Academy. Local organizing committees coordinate with host universities and conference centers like those used by University of Washington, University of Pennsylvania, City University of Hong Kong, and event partners such as IEEE and ACM for logistics, sponsorship, and publication via proceedings aligned with the ACL Anthology.

EMNLP has been held at diverse venues including campuses and convention centers in cities associated with institutions such as Barcelona, Seattle, Doha, Singapore, Hong Kong, Lisbon, Brussels, Montréal, and Hyderabad. Notable editions have coincided with major gatherings where presenters from MIT Media Lab, Max Planck Institute for Informatics, Zhejiang University, Tsinghua University, and Peking University showcased work on corpora like Penn Treebank, OntoNotes, WordNet, and multilingual resources produced in collaboration with European Commission-funded projects. Special workshops and tutorials often link with thematic events such as those organized by NeurIPS, ICLR, and regional ACL chapters including NAACL and EACL.

EMNLP has published influential papers that shaped modern NLP, including empirical evaluations of sequence-to-sequence models inspired by work at Google Brain and OpenAI, and transformer-based innovations that paralleled research from Google Research and Microsoft Research. Landmark contributions have come from teams at Stanford University (e.g., pertaining to natural language understanding datasets), Allen Institute for AI (e.g., entailment and commonsense corpora), Carnegie Mellon University (e.g., parsing and dialog systems), and industrial labs such as Facebook AI Research and DeepMind (e.g., pretrained contextual representations). The conference catalyzed adoption of benchmarks like GLUE and successor suites spearheaded by researchers at New York University and University of Washington, and influenced commercial deployments at Google Assistant, Amazon Web Services, Microsoft Azure, and startups incubated in Silicon Valley and Bangalore.

EMNLP confers best paper and outstanding paper awards selected by program committees with members from University of Chicago, Yale University, Columbia University, Cornell University, Duke University, and leading industry labs. Recognition at EMNLP is cited alongside honors such as the Turing Award-adjacent accolades and community distinctions from the Association for Computational Linguistics. Recipients often include researchers affiliated with University of Edinburgh, Johns Hopkins University, University of Pennsylvania, Massachusetts Institute of Technology, and innovators from Google Research, Microsoft Research, and Facebook AI Research.

Submissions to EMNLP follow guidelines established by the Association for Computational Linguistics with anonymized review processes managed via systems used by venues like NeurIPS and ICML. Program committees draw reviewers from universities such as Stanford University, University of Oxford, Carnegie Mellon University, University of Toronto, and industry researchers at DeepMind, OpenAI, Google Research, and Microsoft Research. Accepted papers are published in the ACL Anthology and presented in oral sessions, poster sessions, and invited talks by scholars from University of Cambridge, ETH Zurich, Tsinghua University, Peking University, and research groups at Facebook AI Research and Google DeepMind.

Category:Natural language processing conferences