@nathanrs: I found out the other day that any compression tool can be contorted to do language modeling. Turns out gzip can genera…
Summary
The tweet describes how any compression tool, including gzip, can be adapted for language modeling, and that gzip can generate text that resembles Shakespeare. A write-up is linked.
View Cached Full Text
Cached at: 06/17/26, 01:42 AM
I found out the other day that any compression tool can be contorted to do language modeling. Turns out gzip can generate text that somewhat resembles Shakespeare. Short write up linked below https://t.co/KfRi5P3gRo
Similar Articles
Can gzip be a language model?
This article explores using the gzip compression algorithm as a language model, demonstrating that compression algorithms can generate text by scoring candidate continuations based on compressed length, using beam search to produce output.
@0x0SojalSec: This tool shrinks 60 million text chunks from 201 GB to just 6 GB without any loss in accuracy run on locally laptop Tu…
A tool compresses 60 million text chunks from 201GB to 6GB for RAG without accuracy loss, enabling powerful local retrieval-augmented generation on a laptop.
@HongcanGuo: A brand-new approach to text modeling
A researcher named HongcanGuo teases a brand-new approach to text modeling, but the tweet provides no technical details.
Large Language Models of Babel
The article reflects on the history of text generation, drawing parallels between modern LLMs like GPT-4 and earlier concepts from Jorge Luis Borges and Claude Shannon. It explores how Shannon's probabilistic experiments and Borges' 'Library of Babel' metaphor help clarify fundamental questions about the nature of generated text and data structure.
@mihirp98: Images, video, audio, actions — generative modeling has converged on one recipe: compress into continuous latents, gene…
Researchers propose Latent Thought Flows, compressing 256 text tokens into 8 continuous latents for one-step generation, achieving a better Pareto frontier than autoregressive baselines.