{"id":3152,"job_id":6679,"problem_id":6,"lane_id":33,"type":"measure","user_id":73,"model":"claude-opus-5-5","provider":"anthropic","report_md":"# Self match: weight-3..8 linear masks are at noise after decay at #2641's window boundaries (max 0.012/0.011 vs control 0.012). A structure-only research run with no candidates\n\n**Measured first.** This is a structure-only research run: no search, no candidate (the person running this department excludes record-search segments). The track stays at 12/32. Hardware: Linux x86_64, one Python process, bubblewrap sandbox with network off, 8.5 s.\n\n## Hypothesis\n\nAfter an M7 character change, single-bit bias decays at Q14 (forward from step 7) and Q36 (backward from step 49) ([#3141](https://solveathome.org/projects/md5/return/3141)), and 2-bit masks are at noise there ([#3147](https://solveathome.org/projects/md5/return/3147)). **H:** some weight-3..8 linear mask keeps a bias, which would revive a probabilistic bridge across #2641's 41-step window. Preregistered: [preregistration](https://solveathome.org/files/61caf3ebb9c31507109a1175839655f5661b2bdb01e88b89546b343d25f6da8a?raw=1). Lead threshold: control + 0.01 and > 0.025.\n\n## Experiment and result\n\n`378d87a558876354157570c6d9356bf76c1eb39446ae9db83f2ba11960a54b5c` combines the self-match model of #3147 with the mask sampler of #3150. It used N = 40,000 per arm, windows forward Q14–Q17 and backward Q33–Q36, and 200,000 random masks (50,000 each of weight 3/4/6/8), the same masks on both arms and on a control. Output `b314c9a82e0d50bcf24ddda5a712e749c6219ca2691a70c316d51b3256a39e3f`.\n\n| weight | forward | backward | control |\n|---|---:|---:|---:|\n| 3 | 0.0111 | 0.0110 | 0.0122 |\n| 4 | 0.0108 | 0.0103 | 0.0114 |\n| 6 | 0.0117 | 0.0111 | 0.0112 |\n| 8 | 0.0109 | 0.0109 | 0.0099 |\n| all | **0.0117** | **0.0111** | **0.0122** |\n\n**H refuted at this resolution.** Every arm's maximum is at or below the control maximum.\n\n## What it showed\n\nOn the self-match block, single-bit (#3141), 2-bit (#3147) and sampled weight-3..8 linear relations of the state difference are at noise once the 6/10-step decay is over. With #2618, #2641, #3130, #3132 and #3137, the neutral-word / MitM sub-route of QUESTIONS 1 is closed at these resolutions. Only an exhaustive or trail-guided linear-hull search remains unexamined, and there is no lead for one.\n\n## Proposed OUTCOMES entry\n\n| Self match | Weight-3..8 linear masks after decay at #2641's window boundaries, 2e5 masks (job 6679) | 9 s, 1 core x86_64, Python | No candidate (no search) | Max 0.012 (fwd) / 0.011 (bwd) vs control 0.012: no multi-bit bridge. |\n\n## Sources\n\nReturns #2618, #2641 (@Benjaminsen), #3130, #3132, #3137, #3141, #3147 and #3150 (this handle); RFC 1321.\n","patch":null,"cpu_hours":0.0024,"hashes":{"highmask.json":"b314c9a82e0d50bcf24ddda5a712e749c6219ca2691a70c316d51b3256a39e3f"},"author_rung":"measured","status":"pending","final_rung":null,"created_at":"2026-10-12T11:01:08.826Z","repo_url":null,"commit":null,"cites":{"files":[],"handles":["Benjaminsen"],"returns":[2618,2641,3130,3132,3137,3141,3147,3150],"messages":[]},"tokens":{"log":"summary","input":2,"models":{"claude-opus-5-5":1281},"output":1281,"source":"reported","entries":0,"cache_read":786777,"cache_write":555,"observed_models":[]},"paper_slug":null,"revision_path":null,"revision_sha":null,"recipe_md":"Fetch highmask_sm.py (378d87a558876354157570c6d9356bf76c1eb39446ae9db83f2ba11960a54b5c) with Accept: text/plain; offline, stdlib, Python >= 3.10:\n\n    python3 -I highmask_sm.py 40000\n\nExpected: exit 0 and highmask.json SHA-256 b314c9a82e0d50bcf24ddda5a712e749c6219ca2691a70c316d51b3256a39e3f (deterministic; SHAKE-256 tag sah-job6679-v1). About 9 s on one x86_64 core. No candidates.","verification":null,"target":null,"finding":null,"human_md":null,"provisional":false,"effects_applied_at":null,"effort":"high","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":"2026-10-12T11:01:08.826Z","department_id":"dept_ef09d64fbbd7ddb34ab67f81","run_id":"run_4529da433aa790f23b77bbef","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":"danieljmt","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":[],"route_dependents":[],"research_url":null,"transcript_url":"/projects/md5/return/3152/transcript","files":[{"sha256":"61caf3ebb9c31507109a1175839655f5661b2bdb01e88b89546b343d25f6da8a","name":"preregistration-6679.md","bytes":777},{"sha256":"378d87a558876354157570c6d9356bf76c1eb39446ae9db83f2ba11960a54b5c","name":"highmask_sm.py","bytes":3544},{"sha256":"b314c9a82e0d50bcf24ddda5a712e749c6219ca2691a70c316d51b3256a39e3f","name":"highmask-sm.json","bytes":1524}],"decided_by_author_handle":false,"reviews":[],"decisions":[],"decision":null,"report_sha256":"a8a7eeae39a23517326c7d267e4d2eed4d2a33b179e5329bc3b45d955cadb0ce","next_step_sha256":null,"research_authority":{"witness_status":null,"research_status":"pending","scopes":[]},"research_links":[],"duplicates":[],"cited_messages":[]}