Do Language Models Know the Way to Rome?

Research output: Chapter in Book/Report/Conference proceeding › Article in proceedings › Research › peer-review

Standard

Do Language Models Know the Way to Rome? / Liétard, Bastien Nathan; Abdou, Mostafa ; Søgaard, Anders.

Proceedings of the Fourth BlackboxNLP Workshop on Analyzing and Interpreting Neural Networks for NLP. Association for Computational Linguistics, 2021. p. 510–517.

Research output: Chapter in Book/Report/Conference proceeding › Article in proceedings › Research › peer-review

Harvard

Liétard, BN, Abdou, M & Søgaard, A 2021, Do Language Models Know the Way to Rome? in Proceedings of the Fourth BlackboxNLP Workshop on Analyzing and Interpreting Neural Networks for NLP. Association for Computational Linguistics, pp. 510–517, Fourth BlackboxNLP Workshop on Analyzing and Interpreting Neural Networks for NLP, Online, 11/11/2021. https://doi.org/10.18653/v1/2021.blackboxnlp-1.40

APA

Liétard, B. N., Abdou, M., & Søgaard, A. (2021). Do Language Models Know the Way to Rome? In Proceedings of the Fourth BlackboxNLP Workshop on Analyzing and Interpreting Neural Networks for NLP (pp. 510–517). Association for Computational Linguistics. https://doi.org/10.18653/v1/2021.blackboxnlp-1.40

Vancouver

Liétard BN, Abdou M, Søgaard A. Do Language Models Know the Way to Rome? In Proceedings of the Fourth BlackboxNLP Workshop on Analyzing and Interpreting Neural Networks for NLP. Association for Computational Linguistics. 2021. p. 510–517 https://doi.org/10.18653/v1/2021.blackboxnlp-1.40

Author

Liétard, Bastien Nathan ; Abdou, Mostafa ; Søgaard, Anders. / Do Language Models Know the Way to Rome?. Proceedings of the Fourth BlackboxNLP Workshop on Analyzing and Interpreting Neural Networks for NLP. Association for Computational Linguistics, 2021. pp. 510–517

Bibtex

@inproceedings{ca13afc3f2e040728b284c5173f8ed2c,
title = "Do Language Models Know the Way to Rome?",
abstract = "The global geometry of language models is important for a range of applications, but language model probes tend to evaluate rather local relations, for which ground truths are easily obtained. In this paper we exploit the fact that in geography, ground truths are available beyond local relations. In a series of experiments, we evaluate the extent to which language model representations of city and country names are isomorphic to real-world geography, e.g., if you tell a language model where Paris and Berlin are, does it know the way to Rome? We find that language models generally encode limited geographic information, but with larger models performing the best, suggesting that geographic knowledge can be induced from higher-order co-occurrence statistics.",
author = "Li{\'e}tard, {Bastien Nathan} and Mostafa Abdou and Anders S{\o}gaard",
year = "2021",
doi = "10.18653/v1/2021.blackboxnlp-1.40",
language = "English",
pages = "510–517",
booktitle = "Proceedings of the Fourth BlackboxNLP Workshop on Analyzing and Interpreting Neural Networks for NLP",
publisher = "Association for Computational Linguistics",
note = "Fourth BlackboxNLP Workshop on Analyzing and Interpreting Neural Networks for NLP ; Conference date: 11-11-2021 Through 11-11-2021",

}

RIS

TY - GEN

T1 - Do Language Models Know the Way to Rome?

AU - Liétard, Bastien Nathan

AU - Abdou, Mostafa

AU - Søgaard, Anders

PY - 2021

Y1 - 2021

N2 - The global geometry of language models is important for a range of applications, but language model probes tend to evaluate rather local relations, for which ground truths are easily obtained. In this paper we exploit the fact that in geography, ground truths are available beyond local relations. In a series of experiments, we evaluate the extent to which language model representations of city and country names are isomorphic to real-world geography, e.g., if you tell a language model where Paris and Berlin are, does it know the way to Rome? We find that language models generally encode limited geographic information, but with larger models performing the best, suggesting that geographic knowledge can be induced from higher-order co-occurrence statistics.

AB - The global geometry of language models is important for a range of applications, but language model probes tend to evaluate rather local relations, for which ground truths are easily obtained. In this paper we exploit the fact that in geography, ground truths are available beyond local relations. In a series of experiments, we evaluate the extent to which language model representations of city and country names are isomorphic to real-world geography, e.g., if you tell a language model where Paris and Berlin are, does it know the way to Rome? We find that language models generally encode limited geographic information, but with larger models performing the best, suggesting that geographic knowledge can be induced from higher-order co-occurrence statistics.

U2 - 10.18653/v1/2021.blackboxnlp-1.40

DO - 10.18653/v1/2021.blackboxnlp-1.40

M3 - Article in proceedings

SP - 510

EP - 517

BT - Proceedings of the Fourth BlackboxNLP Workshop on Analyzing and Interpreting Neural Networks for NLP

PB - Association for Computational Linguistics

T2 - Fourth BlackboxNLP Workshop on Analyzing and Interpreting Neural Networks for NLP

Y2 - 11 November 2021 through 11 November 2021

ER -

ID: 300078921