Too Much Information: Keeping Training Simple for BabyLMs
This work addresses the challenge of efficient language model training for NLP researchers, but it is incremental as it builds on existing BabyLM Challenge efforts with a specific focus on context size.
The paper investigates whether simplifying training data for language models by starting with simpler concepts improves performance, finding that only limiting context size yields benefits, resulting in average gains of 2 points on (Super)GLUE, 1 point on MSGS, and 12% on BLiMP tasks.
This paper details the work of the University of Groningen for the BabyLM Challenge. We follow the idea that, like babies, language models should be introduced to simpler concepts first and build off of that knowledge to understand more complex concepts. We examine this strategy of simple-then-complex through a variety of lenses, namely context size, vocabulary, and overall linguistic complexity of the data. We find that only one, context size, is truly beneficial to training a language model. However this simple change to context size gives us improvements of 2 points on average on (Super)GLUE tasks, 1 point on MSGS tasks, and 12\% on average on BLiMP tasks. Our context-limited model outperforms the baseline that was trained on 10$\times$ the amount of data.