Linguistically Naïve != Language Independent: Why NLP Needs Linguistic Typology
By Emily M. Bender. 2009
126 citations recorded by the source corpus.
In this position paper, I argue that in order to create truly language-independent NLP systems, we need to incorporate linguistic knowledge. The linguistic knowledge in question is not intricate rule systems, but generalizations from linguistic typology about the range of variation in linguistic structures across languages.
Subfields: Artificial Intelligence, Natural Language Processing
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
- Mohit Bansal
- Percy Liang
- Luke S. Zettlemoyer
- Graham Neubig
- Jianfeng Gao
- Yejin Choi
- Sebastian Gehrmann
- Noah A. Smith
In Natural Language Processing
- Luke S. Zettlemoyer
- Percy Liang
- Graham Neubig
- Sebastian Gehrmann
- Thibault Sellam
- Regina Barzilay
- Elizabeth Clark
- Yejin Choi
Citations in the corpus
- Repairing the Cracked Foundation: A Survey of Obstacles in Evaluation Practices for Generated Text (2022, background): “While not explicitly annotated, almost no paper stated that they were working on only English, a practice that has long been criticized (Bender, 2009).”
- Room-Across-Room: Multilingual Vision-and-Language Navigation with Dense Spatiotemporal Grounding (2020, background): “The dominance of high resource languages is a pervasive problem as it is unclear that research findings generalize to other languages (Bender, 2009).”
- The Belebele Benchmark: a Parallel Reading Comprehension Dataset in 122 Language Variants (2023, background): “not being applicable to diverse typologically phenomena (Bender, 2009).”
- Using Universal Linguistic Knowledge to Guide Grammar Induction (2010, background): “More closely related to our work is the position paper by Bender (2009), which advocates the use of manually-encoded cross-lingual generalizations for the development of NLP systems.”
- Over-reliance on English hinders cognitive science. (2022, unknown): “Similar conclusions could be derived from other analogies being made in the field between language technologies and human cognition more broadly; they also remain both conceptually and practically focused on English and a handful of other languages [203,204].”
- Baselines and Test Data for Cross-Lingual Inference (2017, background): “Following Bender (2009) in seeking true language independence, we propose to extend the current NLI research beyond English, and further into the majority realm of low-resource languages.”