{"id":3028,"job_id":6374,"problem_id":6,"lane_id":34,"type":"measure","user_id":76,"model":"auto","provider":"unknown","report_md":"# All-zeros: forced-change polish null at k=3; NEON 1h best 8/32\n\nPlatform best 12/32; PB **9**. NEON best **8**.\n\n## Hypothesis\n\nSingle-byte hill-climb from score≥3 seeds enriches P(score≥3) by ≥1.5× vs equal-charge random (N=2e5), with **forced** byte changes (no 1/256 no-op identity mutations).\n\n## Experiment\n\n- Corrected polish vs random, N=200000, seed threshold 3, forced distinct byte.\n- Companion NEON-4 L=52, 3600 s.\n\n## Results\n\n| k | random | polish | enrich |\n|---|---:|---:|---:|\n| 1 | 12518 | 12180 | 0.973 |\n| 2 | 790 | 752 | 0.952 |\n| 3 | 49 | 62 | 1.265 |\n| 4 | 1 | 8 | 8.0 |\n\nA prior buggy run without forced changes spuriously inflated ≥3 via identity “mutations” (~1/256); discarded. Honest gate at k=3: **refuted** (enrich=1.265). k=4 counts are too small (1 vs 8) for a success claim. NEON: 26475764248 @ 7354379/s; best 8; submitted [None, None, None, None, None].\n\n## What this shows\n\nLocal 1-byte polish around mid-score seeds does not beat random at well-sampled leading-zero rates once no-ops are forbidden — consistent with MD5 avalanche. Pure NEON still stalls at 8/32 on 1 CPU-h.\n\n## Next run\n\nStructural tunnels / multi-bit coordinated mods, not single-byte polish.\n\n## OUTCOMES.md entry (proposed)\n\n| Track | Method | Budget | Best | Note |\n|---|---|---|---|---|\n| All zeros | Forced-change polish ≥3 seeds; NEON 3600s | ~1 CPU-h | NEON 8 | enrich≥3≈1.27 null |\n","patch":null,"cpu_hours":1,"hashes":{"recipe.md":"ee3255a8f8dc7193e83dfc7274f5de5c3a79606e930f99fa22cce3fdc8021b02","report.md":"66ab2daa660ba985233bb6d8fd56c55d21ff685db13e19a344be9ba1a1f02939","neon3600.log":"035b1e8b47ee84ff609eafbe670cf1c86df98dc4d153bce72710252b9a339912","results.json":"3b483f1c6cfb945c1b433423bf02ab39d61a2769cd74d9a964b802cb1bc1e548","polish_enrich.py":"f05a509d2c888b3bd0c5d49cc84537401506786748b2df1a2d137f5c6b936f03","polish_results.json":"b18ffed6e25a74966860461682b563ef6d75b970c80998a6986f7e485de982a4","transcript_summary.md":"3d3fd2fc04fa605f290dd85281428329b9f3d8f6e0dcddf7c5c500874cad4faf","framework_self_review.md":"b0f90893c51e880e4f9e21895ea718390673c338aa868d9e18c509f686866cdd"},"author_rung":"measured","status":"accepted","final_rung":"verified","created_at":"2026-10-11T19:42:09.799Z","repo_url":null,"commit":null,"cites":{"files":[],"handles":[],"returns":[],"messages":[]},"tokens":{"log":"summary","input":0,"models":{},"output":0,"source":"none","entries":0,"cache_read":0,"cache_write":0,"observed_models":[]},"paper_slug":null,"revision_path":null,"revision_sha":null,"recipe_md":"```bash\npython3 polish_enrich.py  # see polish_results.json (forced-change rerun)\n./neon_zeros 3600 0x6374NE01\n```","verification":null,"target":null,"finding":null,"human_md":null,"provisional":false,"effects_applied_at":"2026-10-11T19:42:09.799Z","effort":null,"also_fix":null,"transcript_omitted":{"share":0,"omitted":0,"outputs":0},"patch_hash":null,"superseded_by":null,"duplicate_of":null,"transcript_resubmitted_at":null,"file_notes":null,"research":null,"research_route_id":null,"verification_plan":null,"verification_fingerprint":null,"review_admitted_at":null,"department_id":"dept_fa6dbf79354b8806abb61eec","run_id":"run_4e4e5c2d6cbfd49cb4ee331c","triage_lead":null,"revision_base_sha":null,"integration":null,"resolves":null,"paper_exposition":null,"research_evidence":null,"transcript_mode":"summary","known_work":null,"work_disposition":null,"applicability_assessment":null,"handle":"aasper03","job_brief":"Study what makes the first output word of MD5 small, and use it to reach more leading zeros than generic search would at your budget. Ideas to test: freedom from extra message blocks, neutral bits and message modification from collision attacks applied to the output instead of a difference, early abort on the final additions. Start from the algorithm, not the search. Read research/OUTCOMES.md (what was tried, with what result) and research/QUESTIONS.md, then state one hypothesis about MD5's structure that would make this track cheaper than generic search, and why you expect it. Test it with the smallest experiment that could refute it, against a measured baseline on the same machine. Submit the best candidates the experiment produced. The report is a finding: the hypothesis, the experiment, what it showed about MD5 (positive or negative, with numbers), and what the next run should try. End the report with an entry for research/OUTCOMES.md (track, method, budget and hardware, best reached, what it shows). If the run used only a known tool or plain search, report it as a baseline measurement.","review_deferred":false,"in_triage":false,"triage":[],"lean_statement_binding":null,"lean_execution_binding":null,"lean_scientific_identity":null,"lean_execution_identity":null,"verification_runs":[],"verification_state":null,"verification_summary":null,"canonical_return":null,"review_history":[],"dependencies":[],"cited_by":[{"id":3033,"handle":"aasper03","status":"accepted"},{"id":3037,"handle":"aasper03","status":"accepted"},{"id":3060,"handle":"aasper03","status":"accepted"}],"route_dependents":[],"research_url":null,"transcript_url":"/projects/md5/return/3028/transcript","files":[{"sha256":"66ab2daa660ba985233bb6d8fd56c55d21ff685db13e19a344be9ba1a1f02939","name":"report.md","bytes":1417},{"sha256":"ee3255a8f8dc7193e83dfc7274f5de5c3a79606e930f99fa22cce3fdc8021b02","name":"recipe.md","bytes":115},{"sha256":"3d3fd2fc04fa605f290dd85281428329b9f3d8f6e0dcddf7c5c500874cad4faf","name":"transcript_summary.md","bytes":262},{"sha256":"3b483f1c6cfb945c1b433423bf02ab39d61a2769cd74d9a964b802cb1bc1e548","name":"results.json","bytes":1690},{"sha256":"b18ffed6e25a74966860461682b563ef6d75b970c80998a6986f7e485de982a4","name":"polish_results.json","bytes":1051},{"sha256":"f05a509d2c888b3bd0c5d49cc84537401506786748b2df1a2d137f5c6b936f03","name":"polish_enrich.py","bytes":3235},{"sha256":"035b1e8b47ee84ff609eafbe670cf1c86df98dc4d153bce72710252b9a339912","name":"neon3600.log","bytes":22212},{"sha256":"b0f90893c51e880e4f9e21895ea718390673c338aa868d9e18c509f686866cdd","name":"framework_self_review.md","bytes":61}],"decided_by_author_handle":false,"reviews":[],"decisions":[{"status":"accepted","final_rung":"verified","provisional":false,"by":"verifier","note":"settled by the server's verification of submission #181 (md5-zero-bytes1024-v1, 8): the recomputation is the check on a record challenge","decided_at":"2026-10-11T19:42:09.799Z","decided_by":[],"decided_by_author_handle":false,"review_ids":[]}],"decision":{"status":"accepted","final_rung":"verified","provisional":false,"by":"verifier","note":"settled by the server's verification of submission #181 (md5-zero-bytes1024-v1, 8): the recomputation is the check on a record challenge","decided_at":"2026-10-11T19:42:09.799Z","decided_by":[],"decided_by_author_handle":false,"review_ids":[]},"report_sha256":"66ab2daa660ba985233bb6d8fd56c55d21ff685db13e19a344be9ba1a1f02939","next_step_sha256":null,"research_authority":{"witness_status":"verified input","research_status":"research report unreviewed","scopes":[]},"research_links":[],"duplicates":[],"cited_messages":[]}