Language acquisition in vector space
Bibliographic Data
| ID | 12859664 |
|---|---|
| Authors | Anders Søgaard (0000-0001-5250-4276, University of Copenhagen, corresponding author) |
| Year | 2025 |
| Volume | 22 |
| Issue | 2 |
| Pages | 197-202 |
| Publication date | 2025-04-01 |
| Peer Reviewed | Yes |
| Open Access | No |
| Type | ARTICLE |
| Venue | Intercultural Pragmatics (JOURNAL) |
| Journal identifiers | ISSN: 1612-295X • E-ISSN: 1613-365X |
| Publisher | De Gruyter (PUBLISHER • DE) |
| DOI | 10.1515/ip-2025-2001 |
| OpenAlex | W4412888467 |
| Language | EN |
| Citations received | 1 |
| References cited | 12 |
Language models are mathematical functions and, as such, induce vector spaces in which input is embedded. Comparing the point clouds of concept vectors across such language models and similar computer vision models, we see surprising similarities. This sheds new light on the Innateness Debate. Much linguistic structure can be induced from extra-linguistic data. Language models are generally thought to be too sample-inefficient to be good models of language acquisition, but what about language models initialized by computer vision models
Linguistics · Communication · Computability, Logic, AI Algorithms · Computer Science · Machine Learning and Algorithms · Natural Language Processing Techniques · Philosophy · Psychology
| Unique citing works | 1 |
|---|---|
| Citations per year | 1 |
| Citation span | 2026 - 2026 (1) |
| Citation velocity | current |
| Highly cited | No |
| Citation types | Neutral: 1 |