Sakana AI and UC Berkeley propose MASS self-improvement method
What happened
On 13 October 2026, The Decoder reported that researchers at Sakana AI and UC Berkeley have proposed Multi-Agent Self-Supervision (MASS). The method has a single language model take on three roles at once — solving problems, optimising workflows and evaluating results — to achieve recursive self-improvement. The report also notes that because the same model fills all three roles, its shared blind spots could be a risk. So far this is the only report on the matter, with no follow-up developments or contradictions from other reports.
Written by AI from the reports · updated 4 d ago
Timeline
Follow the reports to see the event from each side.
- The DecoderSakana 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.
Heat of this event
Current heat 3·Peak in the comparable range 10（Oct 13 03:00）·Change over 24 hours in the comparable range -49%
The trend compares only the same accounts observed without a break, so its range can be narrower than the current heat count. Move the pointer or tap the chart to see each hour; with a keyboard, use the left and right arrow keys.