Pre-training LLMs using human-like development data corpus
This work addresses the problem of comparing LLM pre-training to human language acquisition for researchers, but it is incremental as it focuses on replicating and extending existing baselines.
The authors pre-trained LLMs on a human-like development data corpus with token counts similar to those seen by children, establishing baselines and evaluating performance across epochs for the BabyLM shared task.
Pre-trained Large Language Models (LLMs) have shown success in a diverse set of language inference and understanding tasks. The pre-training stage of LLMs looks at a large corpus of raw textual data. The BabyLM shared task compares LLM pre-training to human language acquisition, where the number of tokens seen by 13-year-old kids is magnitudes smaller than the number of tokens seen by LLMs. In this work, we pre-train and evaluate LLMs on their ability to learn contextual word representations using roughly the same number of tokens as seen by children. We provide a strong set of baselines; with different architectures, evaluation of changes in performance across epochs, and reported pre-training metrics for the strict small and strict tracks of the task. We also try to loosely replicate the RoBERTa baseline given by the task organizers to observe the training robustness to hyperparameter selection and replicability. We provide the submission details to the strict and strict-small tracks in this report.