mlsanityData to 7 Oct 2026

Large Language Models are not Models of Natural Language: they are Corpus Models

arXiv 2112.07055 ↗First posted 13 Dec 2021cs.CL cs.LGcited by 6

Not on the map. Why?

Abstract (arXiv)

Natural Language Processing (NLP) has become one of the leading application areas in the current Artificial Intelligence boom. Transfer learning has enabled large deep learning neural networks trained on the language modeling task to vastly improve performance in almost all downstream language tasks. Interestingly, when the language models are trained with data that includes software code, they demonstrate remarkable abilities in generating functioning computer code from natural language specifications. We argue that this creates a conundrum for the claim that eliminative neural models are a radical restructuring in our understanding of cognition in that they eliminate the need for symbolic abstractions like generative phrase structure grammars. Because the syntax of programming languages is by design determined by phrase structure grammars, neural models that produce syntactic code are apparently uninformative about the theoretical foundations of programming languages. The demonstration that neural models perform well on tasks that involve clearly symbolic systems, proves that they cannot be used as an argument that language and other cognitive systems are not symbolic. Finally, we argue as a corollary that the term language model is misleading and propose the adoption of the working term corpus model instead, which better reflects the genesis and contents of the model.

Key facts. Cited by 6 papers in the map (data through 7 Oct 2026). Builds on 17 of its 65 references that are in the map. First posted on arXiv 13 Dec 2021.

Builds on

17 of 65 references.

References · 17

  1. 2017Attention Is All You Needcited by 59,932
  2. 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,204
  3. 2020Language Models are Few-Shot Learnerscited by 32,199
  4. 2021Evaluating Large Language Models Trained on Codecited by 8,053
  5. 2021On the Opportunities and Risks of Foundation Modelscited by 3,961
  6. 2021Program Synthesis with Large Language Modelscited by 3,509
  7. 2020The Pile: An 800GB Dataset of Diverse Text for Language Modelingcited by 2,346
  8. 2019BERT Rediscovers the Classical NLP Pipelinecited by 1,046
  9. 2021Measuring Coding Challenge Competence With APPScited by 879
  10. 2020A Primer in BERTology: What we know about how BERT workscited by 807
  11. 2021Deduplicating Training Data Makes Language Models Bettercited by 705
  12. 2019CCNet: Extracting High Quality Monolingual Datasets from Web Crawl Datacited by 544
5 more
  1. 2019Are Transformers universal approximators of sequence-to-sequence functions?cited by 404
  2. 2018Learning Word Vectors for 157 Languagescited by 279
  3. 2019Assessing BERT's Syntactic Abilitiescited by 191
  4. 2020Learning to Execute Programs with Instruction Pointer Attention Graph Neural Networkscited by 35
  5. 2019Correlating neural and symbolic representations of languagecited by 32

Built on by

Cited by year

Citing papers per year: 2022: 1; 2023: 3; 2024: 2.

6 papers.

  1. 2023Position: Key Claims in LLM Research Have a Long Tail of Footnotescited by 20
  2. 2023A Sentence is Worth a Thousand Pictures: Can Large Language Models Understand Hum4n L4ngu4ge and the W0rld behind W0rds?cited by 17
  3. 2024A Review of the Challenges with Massive Web-mined Corpora Used in Large Language Models Pre-Trainingcited by 7
  4. 2024Evaluating Large Language Models on the GMAT: Implications for the Future of Business Educationcited by 3
  5. 2022The Linguistic Blind Spot of Value-Aligned Agency, Natural and Artificialcited by 1
  6. 2023The Quo Vadis of the Relationship between Language and Large Language Modelscited by 1