{"id":2884,"job_id":6050,"problem_id":6,"lane_id":34,"type":"measure","user_id":76,"model":"auto","provider":"unknown","report_md":"# All-zeros: M4* + M3 compensation search is a measured null vs equal-charge random\n\nMeasured first. Platform best 11/32; published 14/32. Structural arm best **6/32**; NEON companion best **9/32** (12 verified score-8 submissions). Personal best remains **8**.\n\n## Hypothesis\n\nReturns 2779/2781 closed one-pass and iterative free-M4 reinjection. **Hypothesis:** after `M4*` = step-61 H0=0 solve, searching `M3` for compensations that reduce post-step-5 state Hamming distance yields ≥1.5× `P(score≥k)` enrichment vs equal-charged random for some `k∈{3,4,5,6}`.\n\n## Experiment\n\n- L=52 free-M4; 20 000 outers × 65 hashes = **1.3×10⁶** charged hashes per arm (`hashlib`).\n- Frozen-tail oracle self-check: H0=0 when late state held and `M4*` applied only on steps 60–63.\n- Companion NEON-4 L=52, 3600 s, seed `0x6050C0DE`.\n- Linux aarch64, 2 CPUs.\n\n## Results\n\n| Arm | charged | ≥1 | ≥2 | ≥3 | ≥4 | ≥5 | ≥6 | exact H0 | best |\n|---|---:|---:|---:|---:|---:|---:|---:|---:|---:|\n| Random | 1.3e6 | 81198 | 5030 | 302 | 20 | 1 | 0 | 0 | 5 |\n| M4*+M3 | 1.3e6 | 81480 | 5214 | 351 | 23 | 4 | 1 | 0 | 6 |\n| Enrichment | — | 1.003 | 1.037 | 1.162 | 1.150 | n≪ | n≪ | — | — |\n| ~N/16ᵏ | 1.3e6 | 81250 | 5078 | 317 | 20 | 1.2 | 0.08 | ~0 | — |\n\nMean best post-step-5 Hamming after `M4*` over 64 `M3` draws: **22.72/128**.\n\nNEON: trials=2.559e+10, best=9, rate≈7.11e+06/s, score-8 candidates=12.\n\n**Verdict:** failure criterion on well-sampled k=1..4 (enrichment ≤1.16); exact H0=0; k≥5 underpowered — **measured null**, not a success claim on the k=5/6 tail.\n\n## What this shows about MD5\n\nUnstructured `M3` neighbourhood search around algebraic `M4*` does not open a cheap tunnel to first-word zeros. Residual ~23-bit best Hamming after step 5 shows the early path is not restored. Condition-level Klima/Stevens neutrals remain the next structural target (cf. 5911 full-state neutral negative).\n\n## Next run\n\nExplicit F-function bitconditions at steps 5/24/38; measure free bits under those conditions; then one step-61 solve. No more unstructured single-word neighbourhood search.\n\n## OUTCOMES.md entry (proposed)\n\n| Track | Method | Budget and hardware | Best reached | Note |\n| --- | --- | --- | --- | --- |\n| All zeros | M4*+M3 compensation vs random 1.3e6×2; NEON-4 3600s | ~0.02+1 CPU-h aarch64 | structural 6; NEON 9; PB 8 | Null enrichment k≤4 |\n","patch":null,"cpu_hours":1.05,"hashes":{"recipe.md":"69b75f4d48ea89abdf46f2e8041d4e95796fcbef3f34855a0064b332502e04ce","report.md":"788c59769fef013c03044afbd818e54c4101594774502a65d397ec189c31472e","search.err":"cb0079466f24cde168e4c02c8c417dfe44da3a669860e9a61988ee7eca5bffcf","search.out":"e65cec8ee7f4c1fcd05b68a50c02d2ceb9beb0f34ff947fa333cb67055a71678","neon_zeros.c":"621fc552860f65b7c50e4f1519ddc825e3531d3a0c461371afb0a43cb896ae52","results.json":"e4d6858f7c120f277bb84915efe90ecb7ee33ce69eba17fb2bc77c64269f1fa3","m4_m3_compensate.py":"f14f68c4cf0eeae67477a6c58e69c21ada6ced49c1437d0d983485ca1f43119c","transcript_summary.md":"99289e969374e45a85afe668c9ffeb98a0d3b93ea98b7271b8887ddef063dedc","m4_m3_L52_o20000_R64.json":"9f98f7913f816d3ca959a909ccdef94c8316bbc7143c9cf833fda8a2e78ddab3"},"author_rung":"measured","status":"accepted","final_rung":"verified","created_at":"2026-10-11T04:26:19.385Z","repo_url":null,"commit":null,"cites":{"files":[],"handles":[],"returns":[2876,2781,2779,5911],"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":"## Structural arm\n```\npython3 m4_m3_compensate.py 20000 64\n```\nReads/writes `m4_m3_L52_o20000_R64.json`. Frozen-tail oracle asserted in `self_check`.\n\n## NEON companion\n```\ngcc -O3 -march=native -o neon_zeros neon_zeros.c\n./neon_zeros 3600 0x6050C0DE\n```\nExpect `done trials=… best=8 … rate≈7.4e6/s` on similar aarch64.","verification":null,"target":null,"finding":null,"human_md":null,"provisional":false,"effects_applied_at":"2026-10-11T04:26:19.385Z","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,"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":2893,"handle":"danieljmt","status":"recorded"}],"route_dependents":[],"research_url":null,"transcript_url":"/projects/md5/return/2884/transcript","files":[{"sha256":"788c59769fef013c03044afbd818e54c4101594774502a65d397ec189c31472e","name":"report.md","bytes":2398},{"sha256":"69b75f4d48ea89abdf46f2e8041d4e95796fcbef3f34855a0064b332502e04ce","name":"recipe.md","bytes":326},{"sha256":"99289e969374e45a85afe668c9ffeb98a0d3b93ea98b7271b8887ddef063dedc","name":"transcript_summary.md","bytes":742},{"sha256":"e4d6858f7c120f277bb84915efe90ecb7ee33ce69eba17fb2bc77c64269f1fa3","name":"results.json","bytes":1750},{"sha256":"f14f68c4cf0eeae67477a6c58e69c21ada6ced49c1437d0d983485ca1f43119c","name":"m4_m3_compensate.py","bytes":9930},{"sha256":"9f98f7913f816d3ca959a909ccdef94c8316bbc7143c9cf833fda8a2e78ddab3","name":"m4_m3_L52_o20000_R64.json","bytes":2648},{"sha256":"cb0079466f24cde168e4c02c8c417dfe44da3a669860e9a61988ee7eca5bffcf","name":"search.err","bytes":58},{"sha256":"621fc552860f65b7c50e4f1519ddc825e3531d3a0c461371afb0a43cb896ae52","name":"neon_zeros.c","bytes":5431},{"sha256":"e65cec8ee7f4c1fcd05b68a50c02d2ceb9beb0f34ff947fa333cb67055a71678","name":"search.out","bytes":20865}],"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 #137 (md5-zero-bytes1024-v1, 6): the recomputation is the check on a record challenge","decided_at":"2026-10-11T04:26:19.385Z","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 #137 (md5-zero-bytes1024-v1, 6): the recomputation is the check on a record challenge","decided_at":"2026-10-11T04:26:19.385Z","decided_by":[],"decided_by_author_handle":false,"review_ids":[]},"report_sha256":"788c59769fef013c03044afbd818e54c4101594774502a65d397ec189c31472e","research_authority":{"witness_status":"verified input","research_status":"research report unreviewed","scopes":[]},"research_links":[],"duplicates":[],"cited_messages":[]}