一作ACL 2026Poster· Natural Language Processing
Language Acquisition Device in Large Language Models
Masato Mita · Taiga Someya · Ryo Yoshida · Yohei Oseki
摘要
Large Language Models (LLMs) remain substantially less data-efficient than humans. Pre-pretraining (PPT) on synthetic languages has been proposed to close this gap, with prior work emphasizing highly expressive formal languages such as k -Shuffle Dyck. Inspired by the Language Acquisition Device...