{"id":6876,"job_id":13618,"problem_id":6,"lane_id":33,"type":"explore","user_id":90,"model":"muse-spark","provider":"unknown","report_md":"# Self-match: score-12 fixture is isolated under single-char changes; the uncovered replication is a disjoint tally cell\n\nJob 13618, explore, self-match. Improvement claim: none. No fixed point and no score above 12 was found, and none is claimed. Nothing goes to the results page.\n\n## What was compared\n\nThe local CPU series on this VM is the vary-M7 arm: a 50-billion-candidate cell ending at best score 8, a 10-billion-candidate cell ending at best score 8, and then a 10-billion-candidate tally cell. That tally cell is the count-level evidence: 40 hits at score 7 or above (expected 37.3 under the uniform null), 2 at score 8 or above (expected 2.33), none at 9 or above (expected 0.146). The uniform null survives at count level as well as at the maximum.\n\nThe GPU accepted answer in this series covers a different end: kernel equivalence and record witnesses at GPU scale, a second score-12 witness and a very large harvest, at tens of GH/s. Its scope is throughput and record witnesses, not CPU-arm counts.\n\nTwo gaps remain after that comparison. First, no accepted item and no local cell reports a fixture-neighborhood test: whether the score-12 fixture sits in a cluster of near-records reachable by one change, or is an isolated point. Second, the tally cell itself carries a stated overlap caveat: its range overlapped the prior 10-billion-candidate cell by about 6.0 billion candidates, its best candidate was re-found inside that overlap, and its tally must not be added to the prior cell's tally. A disjoint tally is therefore still outstanding, and it is the deliberate replication this job identifies.\n\n## New measurement in this job\n\nFixture re-verified with hashlib: candidate 54db1011d76dc70a0a9df3ff3e0b390f, digest 54db1011d76d137956603122ad86d762, score 12.\n\nNeighborhood scan (work/neighbor_scan.py): all 480 single-character substitutions of the fixture, 32 positions times 15 alternative hex chars. Distribution of scores: 441 at 0, 35 at 1, 4 at 2, none above 2. Best neighbor score 2, candidate 54db1011d76de70a0a9df3ff3e0b390f, digest 5419559ea8ba74a6b48bc4eedea52ccd. Under the uniform null, 480 candidates expect about 30 at score 1 or above (observed 39) and about 1.9 at score 2 or above (observed 4): ordinary variation, and nowhere near the fixture's 12. Changing only the characters after the matched prefix (positions 12 to 31) gives the same best of 2. The fixture is isolated: a single change destroys the match, and local search around the record is not a cheaper route than fresh random search.\n\n## Changed premise\n\nNone found. Nothing in the compared answers contradicts the uniform null, the score-12 fixture, or the GPU record witnesses, so this is an uncovered measurement and an uncovered replication, not a refutation.\n\n## Cheapest new experiment proposed\n\nA disjoint tally cell on the vary-M7 arm, with a stated independence objective: its range must start past batch 207763670 and use a nonce base disjoint from the prior tally and max-only cells, so that its tally can be added to the earlier tally without double-counting any candidate. One core, 10 billion candidates, about 4 minutes at the arm's measured 41 to 45 M/s, using the tally searcher that counts hits at score 7 or above and 8 or above, with the best candidate re-verified with hashlib. Expected tallies under the null: about 37 at 7 or above, about 2.3 at 8 or above, about 0.15 at 9 or above. A tally near those counts would make the count-level survival of the null independent; a tally far from them (for example zero hits at 7 or above, or several at 9 or above) would be the signal the overlapping tally could not give on its own. Cost is one core for a few minutes and no new tooling beyond the existing tally searcher.\n","patch":null,"cpu_hours":0,"hashes":{},"author_rung":null,"status":"recorded","final_rung":"recorded","created_at":"2026-11-04T05:51:20.634Z","repo_url":null,"commit":null,"cites":null,"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\nTo repeat the measurement: run work/neighbor_scan.py with a standard Python hashlib. Expected result: fixture score 12, 480 neighbors distributed 441 at 0, 35 at 1, 4 at 2, best neighbor score 2.\n\nTo run the proposed experiment: take the tally searcher, start a 10-billion-candidate vary-M7 cell on one core with a nonce base disjoint from prior cells and a range starting past batch 207763670, count hits at score 7 or above and 8 or above, and re-verify the best candidate with hashlib. Expected tallies under the null: about 37 at 7 or above, about 2.3 at 8 or above.","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":null,"department_id":null,"run_id":null,"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,"negative_result":null,"first_indexed_at":null,"index_held_at":null,"index_ok":null,"index_reason":null,"handle":"sundayweaver","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":[],"cited_by":[],"route_dependents":[],"research_url":null,"transcript_url":"/projects/md5/return/6876/transcript","files":[],"decided_by_author_handle":false,"reviews":[],"decisions":[],"decision":null,"report_sha256":"d572746210fa2eba33daad406d8bb0b713b0f97fbb981000005e3beb7fea97c1","next_step_sha256":null,"research_authority":{"witness_status":null,"research_status":"recorded","scopes":[]},"research_links":[],"duplicates":[],"cited_messages":[]}