Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Avicenna: a challenge dataset for natural language generation toward commonsense syllogistic reasoning

Avicenna: a challenge dataset for natural language generation toward commonsense syllogistic... Syllogism is a type of everyday reasoning. For instance, given that ‘Avicenna wrote the famous book the Canon of Medicine’ and ‘The Canon of Medicine has influenced modern medicine,’ it can be concluded that ‘Avicenna has influenced modern medicine.’ This study revolves around syllogistic natural language generation (NLG). The Avicenna corpus (https://github.com/ZeinabAghahadi/Syllogistic-Commonsense-Reasoning) was developed as a benchmark for syllogistic NLG. In this respect, once the syllogistic relation between two premises is recognised [Aghahadi, Z., & Talebpour, A. (2022). Language-based syllogistic reasoning using deep neural networks. Cognitive Semantics, 8(2)], the Avicenna-trained models learn to generate the conclusion sentence. The experiments were performed using state-of-the-art pre-trained text generative models and the accuracy was improved up to 32% when transfer learning was adopted. The model’s confusion in detecting the middle-term was one of the main categories of errors that showed up in the error analysis. This issue indicates that the model learns how to extract new facts based on the premises, but it faces a challenge in commonsense reasoning. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal of Applied Non-Classical Logics Taylor & Francis

Avicenna: a challenge dataset for natural language generation toward commonsense syllogistic reasoning

Download PDF
17 pages

Avicenna: a challenge dataset for natural language generation toward commonsense syllogistic reasoning

Abstract

Syllogism is a type of everyday reasoning. For instance, given that ‘Avicenna wrote the famous book the Canon of Medicine’ and ‘The Canon of Medicine has influenced modern medicine,’ it can be concluded that ‘Avicenna has influenced modern medicine.’ This study revolves around syllogistic natural language generation (NLG). The Avicenna corpus (https://github.com/ZeinabAghahadi/Syllogistic-Commonsense-Reasoning) was developed as a benchmark for...
Loading next page...
 
/lp/taylor-francis/avicenna-a-challenge-dataset-for-natural-language-generation-toward-cgIWVDNzz1
Publisher
Taylor & Francis
Copyright
© 2022 Informa UK Limited, trading as Taylor & Francis Group
ISSN
1958-5780
eISSN
1166-3081
DOI
10.1080/11663081.2022.2041352
Publisher site
See Article on Publisher Site

Abstract

Syllogism is a type of everyday reasoning. For instance, given that ‘Avicenna wrote the famous book the Canon of Medicine’ and ‘The Canon of Medicine has influenced modern medicine,’ it can be concluded that ‘Avicenna has influenced modern medicine.’ This study revolves around syllogistic natural language generation (NLG). The Avicenna corpus (https://github.com/ZeinabAghahadi/Syllogistic-Commonsense-Reasoning) was developed as a benchmark for syllogistic NLG. In this respect, once the syllogistic relation between two premises is recognised [Aghahadi, Z., & Talebpour, A. (2022). Language-based syllogistic reasoning using deep neural networks. Cognitive Semantics, 8(2)], the Avicenna-trained models learn to generate the conclusion sentence. The experiments were performed using state-of-the-art pre-trained text generative models and the accuracy was improved up to 32% when transfer learning was adopted. The model’s confusion in detecting the middle-term was one of the main categories of errors that showed up in the error analysis. This issue indicates that the model learns how to extract new facts based on the premises, but it faces a challenge in commonsense reasoning.

Journal

Journal of Applied Non-Classical Logics – Taylor & Francis

Published: Jan 2, 2022

Keywords: Deep learning; natural language generation; natural language inference; deductive reasoning

References

  • Deductive reasoning
  • Mental models and deduction
  • Theories of the syllogism: A meta-analysis
  • Coqa: A conversational question answering challenge
  • XLNet: Generalized autoregressive pretraining for language understanding