Update README.md

2026-07-20 12:20:41 +08:00 · 2026-06-08 07:48:19 +08:00
parent 0d22ee6476
commit 826aa911f7
1 changed files with 1 additions and 1 deletions
@@ -1,4 +1,4 @@
-# projected_grpo — vGROUT
+# vGROUT

 **vGROUT** (vector gradient routing): route the GRPO gradient against an
 extracted reward-hacking direction (in the SVD-of-W basis) to reduce the