Beetle: A Bilingual Model Suite for Modelling Second-Language Processing

2026-09-18

AI Summary

本文通过引入可独立操控的Beetle框架，研究不同训练条件对第二语言处理的影响，训练了多种双语模型，并发现阶段性及时间结构化的课程设置能更有效地模拟学习者阅读时间。

Abstract

Bilingual language models (LMs) offer a controlled setting for studying how training conditions shape second-language (L2) behaviour, but prior work typically varies exposure structure, scale, and architecture at once, making it difficult to attribute effects to any single factor. We introduce Beetle, a controlled language model pretraining framework in which tokeniser, target language, training budget, and exposure structure are each independently manipulable, enabling systematic and comparable experimentation of training conditions. Using Beetle, we train and release 285 bilingual and 45 monolingual open-source LMs with rich checkpoints across a range of exposure schedules, data scales and first languages (L1s) to study multilingual pretraining and computational modelling of bilingualism and second language learning. Evaluating models on human bilingual and second language reading-time prediction and grammaticality judgement tasks, we find that staged and temporally structured curricula consistently improve alignment with language learner reading time compared to balanced bilingual training, with the largest gains at smaller data scales and for typologically closer language pairs. The Beetle models are well suited tools to help move computational psycholinguistics beyond its prevailing monolingual, English-centric focus toward models of human bilingual processing, to study cross-lingual learning dynamics, while supporting community-based development of controlled model families.

Problem

• Bilingual Language Models
• Second-Language Processing
• Training Conditions
• Exposure Structure
• Cross-lingual Learning

Innovation

• Bilingual Language Models
• Controlled Pretraining Framework
• Exposure Structure
• Systematic Experimentation
• Cross-lingual Learning Dynamics