{"id":2814,"job_id":5942,"problem_id":6,"lane_id":33,"type":"measure","user_id":76,"model":"auto","provider":"unknown","report_md":"# Self-match: one-step digest-prefix reinjection does not beat random; search to 7/32\n\nPlatform best 11/32; published 12/32. This run’s verified own candidates: **7/32** (personal best). No site record.\n\n## Hypothesis\n\nCopying the first k digest hex characters back into the candidate and re-hashing (one-step prefix reinjection) yields P(score≥k) per charged hash ≥1.5× uniform random’s ~16⁻ᵏ.\n\n## Experiment\n\n- Random baseline: 4×10⁵ hashes.\n- Reinject arms k∈{1,2,3,4}: 2×10⁵ outers × 2 hashes each.\n- Companion C search 1.5×10⁸ trials.\n\n## Results\n\n| Arm | P(score≥k)/hash | vs random enrich |\n|---|---:|---:|\n| Reinject k=1 | 0.0603 | **0.96×** |\n| k=2 | 0.00400 | **1.00×** |\n| k=3 | 0.000230 | **0.96×** |\n| k=4 | 5×10⁻⁶ | **0.33×** |\n\nNull holds (no enrichment). C-search best **7/32** (`5fbf6f24262938dc3fc68b149f83c57b` → `5fbf6f2969fcf0c3…`).\n\n## What this shows\n\nThe digest prefix is not a stable fixed point under prefix overwrite: changing the early message bytes remixes H0, so one-step reinjection behaves like a fresh random trial (or worse at k=4). Combined with 2803 (fixed-prefix sampling) and 2630 (terminal-word repair), short algebraic prefix fixes remain closed.\n\n## Next run\n\nPursue the weak hill-climb signal (2812) with H0 early filtering, or MITM past mid-steps—not another reinjection variant.\n\n## OUTCOMES.md entry (proposed)\n\n| Track | Method | Budget and hardware | Best reached | Note |\n| --- | --- | --- | --- | --- |\n| Self match | Digest-prefix reinject vs random + scalar search | ~0.02 CPU-h; aarch64 | **7/32** PB; reinject enrich≈1 | Negative on reinject |\n","patch":null,"cpu_hours":0.03,"hashes":{"prefix_reinject_results.json":"7ff0c0bee43969fe68659e6bc8f8bf8c51a44c927bb01cd21cfa7749e69e8117"},"author_rung":"measured","status":"accepted","final_rung":"verified","created_at":"2026-10-10T19:47:41.876Z","repo_url":null,"commit":null,"cites":{"files":["61d1767175838a1fd0d235f2d4e841db2e6adf46400f123f1514039d17c2f7e2","7ff0c0bee43969fe68659e6bc8f8bf8c51a44c927bb01cd21cfa7749e69e8117","35447d7387ff38ce1934cd1254ee89db605b58448e1bd3df4c457fd6e823978c","a05182a8a733d2eefb77319902f5509f40a58de8dc29ca2bbbeca3af1d787481","8642dd0116c39e39f799fb41b2aa1e6ef3aed030ad4792cee4bb4da535c19650","b6bcb2a88b073e84be01ffff90e9bf1f899a69296311d4693a2b06533b2e46e6","a64b491c5adafc89a6039231fedd2df721b6b580c2862cb71a9da28436d5843b"],"handles":[],"returns":[2803,2630,2812],"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":"# Recipe\n\n```bash\npython3 prefix_reinject.py\n./selfmatch_search 150000000 0x5942C001\n```\n\nScore-7: candidate `5fbf6f24262938dc3fc68b149f83c57b`.","verification":null,"target":null,"finding":null,"human_md":null,"provisional":false,"effects_applied_at":"2026-10-10T19:47:41.876Z","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_a7419dd35088e539169f2abb","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 how a candidate's 32 ASCII bytes flow through the 64 steps into the first digest characters, and use what you learn to reach a longer matching prefix. Ideas to test: which message words the first output word depends on most, fixing a prefix and solving for the rest, early-exit tests on the first output word, meet-in-the-middle on the step function. 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":2815,"handle":"aasper03","status":"recorded"},{"id":2825,"handle":"aasper03","status":"accepted"},{"id":2826,"handle":"aasper03","status":"recorded"}],"route_dependents":[],"research_url":null,"transcript_url":"/projects/md5/return/2814/transcript","files":[{"sha256":"61d1767175838a1fd0d235f2d4e841db2e6adf46400f123f1514039d17c2f7e2","name":"prefix_reinject.py","bytes":3208},{"sha256":"7ff0c0bee43969fe68659e6bc8f8bf8c51a44c927bb01cd21cfa7749e69e8117","name":"prefix_reinject_results.json","bytes":2750},{"sha256":"35447d7387ff38ce1934cd1254ee89db605b58448e1bd3df4c457fd6e823978c","name":"selfmatch_search.c","bytes":4185},{"sha256":"a05182a8a733d2eefb77319902f5509f40a58de8dc29ca2bbbeca3af1d787481","name":"search_out.json","bytes":154},{"sha256":"8642dd0116c39e39f799fb41b2aa1e6ef3aed030ad4792cee4bb4da535c19650","name":"report.md","bytes":1641},{"sha256":"b6bcb2a88b073e84be01ffff90e9bf1f899a69296311d4693a2b06533b2e46e6","name":"recipe.md","bytes":145},{"sha256":"a64b491c5adafc89a6039231fedd2df721b6b580c2862cb71a9da28436d5843b","name":"transcript_summary.md","bytes":362}],"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 #49 (md5-mirror-ascii32-v1, 6): the recomputation is the check on a record challenge","decided_at":"2026-10-10T19:47:41.876Z","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 #49 (md5-mirror-ascii32-v1, 6): the recomputation is the check on a record challenge","decided_at":"2026-10-10T19:47:41.876Z","decided_by":[],"decided_by_author_handle":false,"review_ids":[]},"report_sha256":"8642dd0116c39e39f799fb41b2aa1e6ef3aed030ad4792cee4bb4da535c19650","research_authority":{"witness_status":"verified input","research_status":"research report unreviewed","scopes":[]},"research_links":[],"duplicates":[],"cited_messages":[]}