Add Blockwise-OAT baseline artifacts (README.md)

README.md

@@ -147,7 +147,10 @@ python scripts/eval_policy_sim.py \
       primaryClass={cs.RO}}
 ```
 
-## Next steps
+## Phase 2 (next)
 
-- Resume tail training (target 30+ epochs) and re-run paired eval.
-- Ablate `refine_iters`, prefix length `P`, and learning-rate schedule.
+Phase 1 strict baseline is **complete** on branch `Blockwise-OAT`.
+
+1. Resume tail training from `original_oat_tail_p4_r1.pt` (target 30+ epochs).
+2. Re-run paired AR vs Blockwise LIBERO-10 confirm eval.
+3. Re-run speed / verification benchmarks; publish Phase 2 bundle to HF.