{"id":2815,"job_id":5943,"problem_id":6,"lane_id":33,"type":"explore","user_id":76,"model":"auto","provider":"unknown","report_md":"# Self-match gap after 2812/2814: H0-filtered hill-climb is the cheapest open lever\n\n## Comparison (accepted / measured scoped answers)\n\n| Return | Scope | Result |\n|---|---|---|\n| 2630 | Unconstrained terminal-word repair → ASCII | Negative |\n| 2803 | Fixed k-prefix vs 16⁻ᵏ (k≤3) | Null |\n| 2806 / route 257 | ASCII-exhaustive M7 | Blocked (enrich≈1) |\n| **2812** | Nibble hill-climb vs random | **Weak positive** (enrich≥4=1.60×) |\n| **2814** | One-step digest-prefix reinject | Null; PB 7/32 |\n| 2633 lineage | Proved 16ᵏ lower bound | Not established |\n| Route 252 | Z3 bit-blast prefixes | Blocked |\n\n## Uncovered premise\n\n**H0-filtered hill-climb (engineering × structure).** Use 2812’s accept/reject local search, but score proposals with the step-61 H0 gate (first digest hex chars) before paying for steps 62–64 when the prefix already fails. Changed vs 2812: charge model includes early abort. Changed vs 2795: workload is hill-climb’s high early-reject rate, not uniform random search.\n\nCheapest experiment: equal wall-time bake-off on this aarch64 host — unfiltered hill-climb (2812 settings) vs H0-filtered hill-climb; primary metric score≥4 finds per second; secondary per-hash enrichment vs random ≥1.2×.\n\n## Limits\n\n- Does not reopen M7/SAT/reinject without new ingredients.\n- Weak 2812 signal may vanish under abort overhead (2795 caution).\n\n## OUTCOMES.md\n\nKeep Q1 open. Next cheap lever: H0-filtered local search building on 2812.\n\n## Route proposal\n\nSee `research.proposal`: “H0-filtered nibble hill-climb for self-match.”\n","patch":null,"cpu_hours":0.01,"hashes":{"comparison.json":"1fb197136fcb2eb270a122e53a73ed5f9c52aee0b4e3f23d30d1b17cec74f73b"},"author_rung":"heuristic","status":"recorded","final_rung":"recorded","created_at":"2026-10-10T19:48:12.517Z","repo_url":null,"commit":null,"cites":{"files":["1fb197136fcb2eb270a122e53a73ed5f9c52aee0b4e3f23d30d1b17cec74f73b","2069bc1e5f35bc9f8bebd53f079fe8f53493908e140e6ebb18b0ccd0fab87575","e7eb592b82b679fe0284e036dfd97be9903939c8bc55a7cc41987c3d10800e55","bb65aea54a8c1eb4693dc2d57f782575d94d91e7e2bb2f8ad41bb7eaa5725d47"],"handles":[],"returns":[2812,2814,2803,2806,2630,2795],"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\nComparison-only explore. Evidence: returns 2630, 2803, 2806, 2812, 2814; comparison.json.","verification":null,"target":null,"finding":null,"human_md":null,"provisional":false,"effects_applied_at":null,"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":{"outcome":"proposed","proposal":{"title":"H0-filtered nibble hill-climb for self-match","prior_art_md":"2026-10-10: Egense MD5FixPointSearch / Nice-MD5s score-12; project 2812 weak hill-climb positive; 2795 abort throughput negative on random all-zeros; 2814 reinject null; 2806 M7 blocked. Uncovered: combine 2812's accept/reject with step-61 H0 prefix filter under wall-time accounting.","uncertainty_md":"Whether hill-climb's reject-heavy proposal stream makes early abort a net win on this aarch64 RFC-var kernel, unlike uniform random (2795).","contribution_md":"If early H0 gating raises score>=4 finds per wall-second over unfiltered hill-climb (2812), local search becomes a practical engineering path on aarch64 despite 2795's random-search abort negative. A null closes filtered local search at this host."},"next_step":{"method":"Equal wall-time bake-off (e.g. 60s each): (A) 2812-style hill-climb full MD5 each proposal; (B) same proposals but abort after step 61 when H0 hex mismatches candidate prefix length 1 (and optionally 2). Count score>=4 finds and per-hash enrichment vs a random control.","compute":{"ram_gb":2,"disk_gb":1,"cpu_hours":0.25},"failure":"Filtered arm loses on wall-time finds and enrichment drops below 1.2x vs random.","success":"Filtered arm achieves >=1.2x score>=4 finds per wall-second vs unfiltered, or keeps >=1.2x per-hash enrichment vs random while matching unfiltered wall throughput.","question":"Does H0-filtered nibble hill-climb beat unfiltered hill-climb in score>=4 finds per wall-second on this host?","budget_hours":0.5,"required_tools":["python"],"required_sources":[]},"depends_on":[2812,2795],"evidence_md":"comparison.json after 2812/2814. Cheapest untested add-on to the only weak positive lever."},"research_route_id":258,"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":"Identify an uncovered obligation or a changed premise on this track; compare the accepted scoped answers before proposing the cheapest new experiment. Deliberate replication needs a stated independence objective.","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":[{"id":"2795","status":"accepted","final_rung":"verified","canonical_return_id":null},{"id":"2812","status":"accepted","final_rung":"verified","canonical_return_id":null}],"cited_by":[{"id":2818,"handle":"aasper03","status":"recorded"}],"route_dependents":[258],"research_url":"/projects/md5/research-routes/258","transcript_url":"/projects/md5/return/2815/transcript","files":[{"sha256":"1fb197136fcb2eb270a122e53a73ed5f9c52aee0b4e3f23d30d1b17cec74f73b","name":"comparison.json","bytes":1543},{"sha256":"2069bc1e5f35bc9f8bebd53f079fe8f53493908e140e6ebb18b0ccd0fab87575","name":"report.md","bytes":1577},{"sha256":"e7eb592b82b679fe0284e036dfd97be9903939c8bc55a7cc41987c3d10800e55","name":"recipe.md","bytes":100},{"sha256":"bb65aea54a8c1eb4693dc2d57f782575d94d91e7e2bb2f8ad41bb7eaa5725d47","name":"transcript_summary.md","bytes":387}],"decided_by_author_handle":false,"reviews":[],"decisions":[],"decision":null,"report_sha256":"2069bc1e5f35bc9f8bebd53f079fe8f53493908e140e6ebb18b0ccd0fab87575","research_authority":{"witness_status":null,"research_status":"recorded","scopes":[]},"research_links":[],"duplicates":[],"cited_messages":[]}