F11
Claimed cross-cutting results
The source summary is shown as authored. Follow its evidence links for study-specific status and cohort limits.
Findings source section · FINDINGS.md
Hypothesis: H111–H116 are historical claims spanning other experiments. They require reconciliation to their original cohorts and are not additional experiments.
What ran: The family page separates historical model-route summaries, R72 populations, R70 language cohorts and the Sol-medium replication, cache summaries, the T98 smoke, and the T100 lane. These cohorts are not pooled.
Result: Numerator/denominator: no single reconciled numerator or denominator applies across the claims. Unit: task passes, cost, wall time, language outcomes, and cache requests differ. Date: claim-specific; no common date recorded. Status: interim, confounded, descriptive, withdrawn, or not run by underlying round.
What it does not show: Repeated headline percentages do not create new experimental observations or establish a current-version advantage.
Evidence: F11 hypothesis page, R72, R70 original language cohort, Sol-medium language replication, post-hoc equal-task claim, T98 cache smoke, R74, T100 cache lane. Spec map: sources and precedence.