From d5006f268d5b465d1d06ca0df0bd823cb7aff4d9 Mon Sep 17 00:00:00 2001
From: Ronald Tse <ronald.tse@gmail.com>
Date: Sun, 18 Oct 2026 17:43:34 +0800
Subject: [PATCH] docs(wO39): stage-1 complete (oracle 0.54%, -25% vs best
 arm); stage-2 launched via spine

---
 TODO.sota/39-ar-oracle-distill.md | 16 ++++++++++++++++
 1 file changed, 16 insertions(+)

diff --git a/TODO.sota/39-ar-oracle-distill.md b/TODO.sota/39-ar-oracle-distill.md
index 9c5444b3..a300c37c 100644
--- a/TODO.sota/39-ar-oracle-distill.md
+++ b/TODO.sota/39-ar-oracle-distill.md
@@ -61,3 +61,19 @@ silver itself.
 
 ≈ 20-28 GPU-h total (relable 10-14 + train 10-14). Under every
 prior arm's budget; no new data, no new infra.
+
+## Stage-1 COMPLETE (2026-10-18, resume9)
+
+290,575 rows consolidated in minutes once the consolidation levenshtein
+moved to the C wheel (the pure-python loop needed ~56h vs the 23h
+timeout — resume6/8 died exactly there). Stats:
+
+- arm CER vs silver: p047 0.72% · p040 0.74% · p029 0.82%
+- disagreement 31.4% — the oracle's opportunity space
+- **oracle CER 0.54%** (−25% vs best single arm); picks: p047 87%,
+  p040 11%, p029 2%
+
+**Stage-2 LAUNCHED THROUGH THE SPINE** (job 6ad494b9): the ara recipe's
+sft stage planned it (O_* envs, both mounts) and the Runner's budget
+guard cleared it — the first dogfooded launch; hand-typed job commands
+retire here. Gate: first-alt WER < 3.5607; stretch ≤ 2.70.