Research

Sakana AI's self-improving model gets better at solving and judging, but can't catch its own blind spots.

The DecoderIndependent reporting·
AI brief

Sakana AI's MASS method lets one language model improve itself by solving tasks, judging results, and refining workflows, reaching a win rate of nearly 70 percent against the base model after two rounds.

Written by AI from The Decoder's published text.

Source

“Sakana AI's MASS method lets one language model improve itself by solving tasks, judging results, and refining workflows, reaching a win rate of nearly 70 percent against the base model after two rounds.”
The Decoder, 12 Oct 2026, 17:58.

Reporting from a publisher outside the organisation being covered.

Read original story ↗