Reasoning with Transformer-based Models: Deep Learning, but Shallow Reasoning - Archive ouverte HAL Access content directly
Conference Papers Year :

Reasoning with Transformer-based Models: Deep Learning, but Shallow Reasoning

(1) , (2, 3, 4) , (2, 5, 6)
1
2
3
4
5
6

Abstract

Recent years have seen impressive performance of transformer-based models on different natural language processing tasks. However, it is not clear to what degree the transformers can reason on natural language. To shed light on this question, this survey paper discusses the performance of transformers on different reasoning tasks, including mathematical reasoning, commonsense reasoning, and logical reasoning. We point out successes and limitations, of both empirical and theoretical nature.
Fichier principal
Vignette du fichier
akbc-2021-reasoning.pdf (335.71 Ko) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

hal-03344668 , version 1 (15-09-2021)

Identifiers

  • HAL Id : hal-03344668 , version 1

Cite

Chadi Helwe, Chloé Clavel, Fabian Suchanek. Reasoning with Transformer-based Models: Deep Learning, but Shallow Reasoning. International Conference on Automated Knowledge Base Construction (AKBC), 2021, online, United States. ⟨hal-03344668⟩
77 View
19 Download

Share

Gmail Facebook Twitter LinkedIn More