pre-register item 16: code->GSM8K cross-task transfer

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-07-15 12:54:58 +02:00
parent b013ca27f1
commit ebf0eb8f49
1 changed files with 13 additions and 0 deletions
@@ -256,3 +256,16 @@ essentially unmoved: mean 0.298, range [0.285, 0.310], 0/1536 dims moved
 hand-tuned 0.3 was already optimal. Recipe consequence: fixed-alpha
 anchored merge is the recommended design; learnable-alpha safe but
 pointless, learnable-B harmful, per-depth strands the gain.
+
+16. **Code→GSM8K cross-task transfer (pre-registered 2026-07-15 ~14:10,
+    before running).** The MBPP-trained loop adapter (adapter_code, s0) and
+    the noise-s0 variant evaluated on GSM8K test (n=256, prompt-only loop,
+    same harness as eval_gsmonly). Extends the transfer-distance ladder
+    (HumanEval tie -> LCB trained-hurts) across tasks. Predictions:
+    (a) hard-bucket gain ~0 (plan content is task-local; GSM8K needs
+    evolving state, not static plans); (b) easy items damaged at k>0
+    (~93 -> 50-70%), comparable to or worse than the GSM-trained merge —
+    substrate damage on GSM8K is perturbation-driven and content-agnostic;
+    (c) overall at k>0 below k=0 (no rescue). If instead hard gains
+    appear (>5 points), plan-shaped content is partially task-general —
+    would weaken the task-local claim from LCB.