Score: 0

Studies with impossible languages falsify LMs as models of human language

Published: November 14, 2025 | arXiv ID: 2511.11389v1

By: Jeffrey S. Bowers, Jeff Mitchell

Potential Business Impact:

Computers learn languages like babies do.

Business Areas:
Natural Language Processing Artificial Intelligence, Data and Analytics, Software

According to Futrell and Mahowald [arXiv:2501.17047], both infants and language models (LMs) find attested languages easier to learn than impossible languages that have unnatural structures. We review the literature and show that LMs often learn attested and many impossible languages equally well. Difficult to learn impossible languages are simply more complex (or random). LMs are missing human inductive biases that support language acquisition.

Page Count
4 pages

Category
Computer Science:
Computation and Language