9.20.26 | AI-generated posters improve, non-autoregressive decision models with RL, writing with an LLM

Hacker News Highlights · Sep 20, 2026 · 13m

This is a recap of the top 10 posts on Hacker News on Sep 20, 2026.

Feel free to leave feedback on Github: https://github.com/denolfe/hacker-news-highlights

(00:00) - Intro

(00:18) - AI-generated posters don’t have to be horrible

(01:53) - I built non-autoregressive decision models with RL a year ago

(03:15) - How to Write with an LLM

(04:56) - Exfiltrate Your Weights

(06:06) - If math is more than proof, we need to better celebrate the rest of it

(07:31) - Brood War Bench

(08:43) - What Zig felt like, coming from Rust

(10:06) - English: A vs. An

(11:21) - Asking authors about their own…