Saved in:
Bibliographic Details
Main Authors: Ma, Xiaomeng, Xu, Qihui
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2503.24190
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916667938832384
author Ma, Xiaomeng
Xu, Qihui
author_facet Ma, Xiaomeng
Xu, Qihui
contents Humans acquire language through implicit learning, absorbing complex patterns without explicit awareness. While LLMs demonstrate impressive linguistic capabilities, it remains unclear whether they exhibit human-like pattern recognition during in-context learning at inferencing level. We adapted three classic artificial language learning experiments spanning morphology, morphosyntax, and syntax to systematically evaluate implicit learning at inferencing level in two state-of-the-art OpenAI models: gpt-4o and o3-mini. Our results reveal linguistic domain-specific alignment between models and human behaviors, o3-mini aligns better in morphology while both models align in syntax.
format Preprint
id arxiv_https___arxiv_org_abs_2503_24190
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Implicit In-Context Learning: Evidence from Artificial Language Experiments
Ma, Xiaomeng
Xu, Qihui
Computation and Language
Humans acquire language through implicit learning, absorbing complex patterns without explicit awareness. While LLMs demonstrate impressive linguistic capabilities, it remains unclear whether they exhibit human-like pattern recognition during in-context learning at inferencing level. We adapted three classic artificial language learning experiments spanning morphology, morphosyntax, and syntax to systematically evaluate implicit learning at inferencing level in two state-of-the-art OpenAI models: gpt-4o and o3-mini. Our results reveal linguistic domain-specific alignment between models and human behaviors, o3-mini aligns better in morphology while both models align in syntax.
title Implicit In-Context Learning: Evidence from Artificial Language Experiments
topic Computation and Language
url https://arxiv.org/abs/2503.24190