Unigram-Normalized Perplexity as a Language Model Performance Measure with Different Vocabulary Sizes

Abstract: Although Perplexity is a widely used performance metric for LLMs, the values are highly dependent upon the number of words in the corpus and is useful to compare performance of the same corpus only. In this paper, we propose a new metric that can be used to evaluate LLM performance with different vocabulary sizes. The proposed unigram-normalized Perplexity actually presents the performance improvement of the LLMs from that of simple unigram model, and is robust on the vocabulary size. Both theoretical analysis and computational experiments are reported.