Can gzip be a language model?

Posted by asasidh 13 hours ago

Counter6Comment1OpenOriginal

Comments

Comment by asasidh 13 hours ago

The author primarily talks about the compression–prediction equivalence and also provides some working code linked in Github https://github.com/nathan-barry/gzipt

Every prediction model is inherently a compressor, and all compression algorithms are prediction models.

Reference: Language Modeling Is Compression — Delétang et al., DeepMind, 2023. The prediction-compression equivalence, with the Chinchilla-beats-PNG result.