Seonglae ChoLanguage Modeling Is Compression
It has long been established that predictive models can be transformed into lossless compressors and vice versa. Incidentally, in recent years, the machine learning community has focused on...
https://arxiv.org/abs/2309.10668
why lowercase letters save data | endtimes.dev
Lowercase letters and uppercase letters use the same amout of data — 1 byte each.
https://endtimes.dev/why-lowercase-letters-save-data/