Dango: A Strictly L1-Only Large Language Model for Studying Second Language Acquisition
Abstract
AbstractWe introduce Dango, a 1.8B-parameter large language model designed for controlled studies of L1→L2 (Japanese→English) transfer in second language acquisition (SLA). While previous studies have explored SLA in language models, they have predominantly relied on smaller or non-decoder models, limiting their ability to generate open-ended text and reducing their suitability as practical L2 simulators. We identify a key challenge when scaling models to this size: L2 contamination within the "monolingual" pretraining corpus used for L1 acquisition. To address this, we propose a filtering method to reduce premature exposure to English while preserving realistic, minimal exposure. We then fine-tune the model on LLM-generated L2-learning lessons to simulate the L2 acquisition process. Our evaluations show that Dango develops human-like L2 production patterns, reliably matching Japanese learners of English more closely than the multilingual baseline across our distributional metrics and the unfiltered baseline on our usage frequency metric. We release the model, data, and code to facilitate reproducible computational SLA research and learner-facing applications.