Evaluating the Language Abilities of Large Language Models vs. Humans: Three Caveats

Evaluating the Language Abilities of Large Language Models vs. Humans: Three Caveats

We identify and analyze three caveats that may arise when analyzing the linguistic abilities of Large Language Models. The problem of unlicensed generalizations refers to the danger of interpreting performance in one task as predictive of the models’ overall capabilities, based on the assumption tha...

Full description

Bibliographic Details
Main Authors: Evelina Leivada, Vittoria Dentella, Fritz Günther
Format: Article
Language:English
Published: PsychOpen GOLD/ Leibniz Institute for Psychology 2024-04-01
Series:Biolinguistics
Subjects:
artificial intelligence
grammaticality
large language models
probabilities
Online Access:https://doi.org/10.5964/bioling.14391

Internet

https://doi.org/10.5964/bioling.14391

Similar Items