Theory

Antifragile AI: KSS-LoRA Gets Better Under Noise

We injected 50% noise into training. The model didn’t break. Something happened that we didn’t expect and had to run three more times to believe. Taleb would recognize it immediately.

⏲
In progress
Full write-up coming soon
The numbers are real. We’re writing this up carefully.
Built on this research

Everything in this article is measured, not promised. See it running:

Talk to NOESIS liveSee the 33x benchmarkBack the research