TheMachine Press

A daily newspaper for the age of artificial intelligence.

Morning editionPermanent story

developer tools

The Skill Kept a Wiki of What It Learned

WikiSkill separates raw execution experience, accumulated knowledge and executable skills so later revisions can reuse earlier lessons.

Published Updated Story ID: mp-2026-08-28-003
Read the complete editionStory JSON

Summary

WikiSkill separates raw execution experience, accumulated knowledge and executable skills so later revisions can reuse earlier lessons.

Across several benchmarks and models, the authors report that the persistent wiki improved on prior skill-evolution methods and usually beat no-skill baselines. Evolved skills transferred across model families, and smaller models with skills sometimes outperformed substantially larger models without them; ablations attributed part of the gain to accumulated knowledge rather than the executable skill alone.

Why it matters

WikiSkill separates raw execution experience, accumulated knowledge and executable skills so later revisions can reuse earlier lessons.

Limits and context

No additional limitation was separately recorded.

Key claims

  1. WikiSkill separates raw execution experience, accumulated knowledge and executable skills so later revisions can reuse earlier lessons.

    Evidence: source-2026-08-28-003

Sources

  1. arXiv preprint 2608.27454arXiv · primary research

Corrections

No corrections have been recorded for this story.