@hi_tysam on Backlist

2 appearances on the backlist front page in the last 30 days.

64.

A little over 2 years ago, I solved the SolidGoldMagikarp stability problem.

A little over 2 years ago, I solved the SolidGoldMagikarp stability problem. Today, I am releasing the results of that work as a new technique to regularize training. More details below.

by @hi_tysam (Fern) · backlist 2026-05-26 · rubric 91.0

76.

Autoregressive transformers have a core problem that limits their decoding performance: teacher forcing. This tec…

Autoregressive transformers have a core problem that limits their decoding performance: teacher forcing. This technique has been around for a while, and has let us train them massively in parallel. But it has a significant inference gap tha

by @hi_tysam (Fern) · backlist 2026-05-26 · rubric 89.0