| … | … | @@ -140,3 +140,4 @@ knowledge-ratchet/ |
| 140 | 140 | ## Inspired By |
| 141 | 141 | |
| 142 | 142 | Karpathy's [autoresearch](https://github.com/karpathy/autoresearch) — automated experiment loops with a forward-only ratchet. Adapted from scalar metric optimization to domain knowledge improvement. The metric isn't val_bpb — it's "did the user have to correct me?" |
| 143 | # trigger |