Enhancing Linguistic Competence of Language Models through Pre-training with Language Learning Tasks
Enhances linguistic competence of LMs for NLP practitioners, with incremental gains over standard pre-training.
L2T integrates language learning tasks into LM pre-training to improve linguistic competence, achieving faster acquisition and competitive reasoning performance.
Language models (LMs) are pre-trained on raw text datasets to generate text sequences token-by-token. While this approach facilitates the learning of world knowledge and reasoning, it does not explicitly optimize for linguistic competence. To bridge this gap, we propose L2T, a pre-training framework integrating Language Learning Tasks alongside standard next-token prediction. Inspired by human language acquisition, L2T transforms raw text into structured input-output pairs to provide explicit linguistic stimulation. Pre-training LMs on a mixture of raw text and L2T data not only improves overall performance on linguistic competence benchmarks but accelerates its acquisition, while maintaining competitive performance on general reasoning tasks.