Commit Graph

4 Commits

Author SHA1 Message Date
8e6290e7e0 docs: finalize deferred at N=5 (median 12.89M, ~tied with orig) 2026-06-16 02:12:12 +00:00
3bf3316572 docs: FINDINGS.md — benchmark synthesis; track raw results data
Capstone summary of the Builder/Adversary prompt + verification-cadence study:
- adversary EXISTENCE costs ~4.7x (solo 2.8M vs ~13M); cadence is ~token-neutral
- context hygiene is the one clean -22% win; minimal prompts -25% but test less
- deferred review saves nothing (the one comprehensive pass is expensive) + late
- cost is process not product (tokens~duration 0.83, ~commits 0.79, ~LOC -0.04)
All results now in-repo: FINDINGS.md + RESULTS-campaign.md + raw .data + runners.
(deferred N=3, finalizing to N=5.)
2026-06-16 01:53:34 +00:00
aeee484395 results: full 5-variant campaign complete (incl. builder-solo control) 2026-06-15 07:42:46 +00:00
29b89140e7 results: 4-variant campaign complete (5/5 each); analysis with ratios 2026-06-15 06:40:34 +00:00