The Decoder

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

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. But when the model misses the same errors across all three roles, it can mistake them for progress and reinforce them during training.

Tarun Mottlia

Via The Decoder

Oct 12, 2026·Excerpt

Sakana AI's self-improving model gets better at solving and judging, but can't catch its own blind spots — Image: Copies of the same model run the workflow as a team, while a self-evaluator compares the result against the current champion. A self-optimizer uses that feedback to write the next workflow, and the model learns from its best runs after each round. | Image: Sakana AI

Sakana AI's MASS method lets one language model improve itself by solving tasks, judging results, and refining workflows without a stronger teacher model. The study finds gains across all three roles, but shared blind spots can let errors pass as progress and feed into the next training cycle.

Researchers from Sakana AI and UC Berkeley have introduced "Multi-Agent Self-Supervision" (MASS), a method for recursive self-improvement (RSI). A single language model handles the task, optimizes how the work gets done, and…

Continue reading on The Decoder ↗

Courtesy

This story was originally published by The Decoder. All rights belong to the original publisher.

Read the original on the-decoder.com ↗

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

Courtesy

Have a story tip, correction or partnership idea?