A Comparison of Elementary Baselines for BabyLM

AbstractThis paper explores multiple simple baselines for the BabyLM challenge, covering random models, elementary predictions based on frequency, n-gram language models, LSTM with several tokenizers (BPE, Unigram, SuperBPE), and GPT-BERT, the winning architecture from the prior BabyLM edition. The evaluation is focused on the BLiMP and BLiMP-Supplement benchmarks. Our experiments show that Strict-Small can sometimes outperform Strict, the fact that performance can be highly sensitive to tokenization and the importance of data efficiency. A simple word-frequency baseline scored unexpectedly high, which led to identifying an evaluation artifact in the pipeline: a system that returns identical logits for both sentences in a minimal pair can achieve maximal accuracy.