I thought that was about LLMs being trained on compressed data. But I might be t... | Hacker News

Hacker Newsnew | past | comments | ask | show | jobs | submit

		jmholla on Aug 16, 2023 \| parent \| context \| favorite \| on: Ts_zip: Text Compression Using Large Language Mode... I thought that was about LLMs being trained on compressed data. But I might be thinking about a different paper.

dragonwriter on Aug 16, 2023 [–]

Gzip + kNN for text classification:

https://aclanthology.org/2023.findings-acl.426.pdf

junipertea on Aug 16, 2023 | [–]

Not LLM, just BERT, also did not actually outperform it.

source: https://kenschutte.com/gzip-knn-paper/

Consider applying for YC's Winter 2026 batch! Applications are open till Nov 10
Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact