ChatGPT Usage and Limitations - Archive ouverte HAL Accéder directement au contenu
Pré-Publication, Document De Travail Année : 2022

ChatGPT Usage and Limitations

Amos Azaria
  • Fonction : Auteur
  • PersonId : 1209784

Résumé

Large language models have been shown useful in multiple domains including conversational agents, education, and explainable AI. ChatGPT is a large language model developed by OpenAI as a conversational agent. Being a large language model, ChatGPT is trained on massive amounts of data. Clearly, the characteristics of the data influence ChatGPT's responses. In this paper, we stress a surprising bias of ChatGPT related to the use of digits in numbers. Namely, we show a very high correlation between the frequency of digits generated by ChatGPT and humans' favorite numbers, with the most frequent digit generated by ChatGPT, matching humans' most favorable number, 7. We also show some advantages of ChatGPT being developed as a conversational agent, and discuss some of its limitations.
Fichier principal
Vignette du fichier
ChatGPT.pdf (252.69 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-03913837 , version 1 (27-12-2022)

Identifiants

  • HAL Id : hal-03913837 , version 1

Citer

Amos Azaria. ChatGPT Usage and Limitations. 2022. ⟨hal-03913837⟩
8949 Consultations
3373 Téléchargements

Partager

Gmail Facebook X LinkedIn More