ys.yoyo’s stash

An independent research journal

Questions, experiments,
and what follows.

A growing collection of research, documented from the first question
to the full experiment. Open methods. Complete results.

Explore the research ↓
FOLLOW THE QUESTION.KEEP THE EVIDENCE.

The research

01 studies in the stash
Find a question

1 study found

01

Machine learning / Research

Learning from reasoning mistakes with 0.1% of the weights

An exploratory Qwen3 study of self-correction, sparse training, and the gap between solving a mistake and learning a skill.

Oct 18, 2026·13 min read Code & checkpoints
✳

More questions to come.

This is a living journal. Each study adds another piece to the picture.

About the stash