{"analysisPlan":[{"method":"jinn.benchmarking.method/wilson","parameters":{"verdictRule":"sole"},"version":"1"},{"method":"jinn.benchmarking.method/binary-instrument","parameters":{"candidateClasses":["correct","specific-wrong","vague-topical-wrong"],"intervalAlpha":"0.05","judgeModelProfile":"dated-snapshot-sampling","k":3,"measurementProfile":"binary-instrument@1","parserInvalidPolicy":"abstain","promptedScreeningProfile":"prompted-codex-screening/v1","reduction":"strict-majority","strata":["category-1","category-2","category-3","category-4"],"truthAdmission":"screened-operator-sampled","verdictRule":"sole"},"version":"1"},{"method":"jinn.benchmarking.method/pairwise-disagreement","parameters":{"candidateClasses":["correct","specific-wrong","vague-topical-wrong"],"intervalAlpha":"0.05","k":3,"measurementProfile":"binary-instrument@1","parserInvalidPolicy":"abstain","reduction":"strict-majority","strata":["category-1","category-2","category-3","category-4"],"truthAdmission":"screened-operator-sampled","verdictRule":"sole"},"version":"1"},{"method":"jinn.benchmarking.method/paired-majority-delta","parameters":{"alpha":"0.05","baseline":"mem0","candidate":"mem0-evidence","candidateClasses":["correct","specific-wrong","vague-topical-wrong"],"k":3,"measurementProfile":"binary-instrument@1","parserInvalidPolicy":"abstain","reduction":"strict-majority","resamples":20000,"seed":20260819,"strata":["category-1","category-2","category-3","category-4"],"truthAdmission":"screened-operator-sampled","verdictRule":"sole"},"version":"1"}],"arms":[{"armId":"audited","pinning":{"harness":{"id":"inspect-ai-judge","version":"1"},"model":{"id":"gpt-4o-mini-2024-07-18"},"network.jinn.binary-judgment.instrument":"sha256:1956c8fe0a7a10b6301e111fe83a6446c612d7371953302b8c77b2d025fc25b5"}},{"armId":"backboard","pinning":{"harness":{"id":"inspect-ai-judge","version":"1"},"model":{"id":"gpt-4o-mini-2024-07-18"},"network.jinn.binary-judgment.instrument":"sha256:4771a07e88e8acb4a084255bf2ba8cbe62e36004601a0a00856d6f0fb0dfee8b"}},{"armId":"mem0","pinning":{"harness":{"id":"inspect-ai-judge","version":"1"},"model":{"id":"gpt-4o-mini-2024-07-18"},"network.jinn.binary-judgment.instrument":"sha256:ed91e8559bc259142a22b26d36126334b97704b48e9711e41ddd3866f6fa4f0f"}},{"armId":"mem0-evidence","pinning":{"harness":{"id":"inspect-ai-judge","version":"1"},"model":{"id":"gpt-4o-mini-2024-07-18"},"network.jinn.binary-judgment.instrument":"sha256:47473dfe08a02038ba35192fb8a430c0c3cf3680545e20c478bd1197d3e04b1a"}},{"armId":"revised","pinning":{"harness":{"id":"inspect-ai-judge","version":"1"},"model":{"id":"gpt-4o-mini-2024-07-18"},"network.jinn.binary-judgment.instrument":"sha256:8b8ae208c62dd4afe0c8cceb58a18b11511158324d106dd7bfac60863ae29476"}},{"armId":"strict-dial","pinning":{"harness":{"id":"inspect-ai-judge","version":"1"},"model":{"id":"gpt-4o-mini-2024-07-18"},"network.jinn.binary-judgment.instrument":"sha256:d7f663de1795438f28c7f7c8a2aa3288d12ad6a3e41186321ce3b1c8ffd417e2"}}],"benchmark":{"digest":{"sha256":"9ae50617f9112b750518c04309b96648207f6d0e17ba044a077d0d5185b84c9e"}},"closeAt":"2026-08-29T16:30:51.384Z","https://spec.jinn.network/extensions/benchmark-publication/v1":{"registrationArtifacts":[{"artifact":{"digest":{"sha256":"12a214a8c8cdf631f8bd83b2fb5d369140ae6823ea5a6cc7b9a31ed6ac20aec8"},"mediaType":"application/json"},"role":"https://product.jinn.network/artifact-roles/inspect/selection-manifest/v1"}]},"owner":"did:key:z6Mkmz8SWiUshwDMszSqwzMngt7ScNmsEj7vZoFRLjpcZTm9","policy":{"cellWindow":3600000,"completenessFloor":"0.995","evaluation":{"distinctEvaluator":false,"maxInfrastructureRetries":1,"minVerdicts":1},"independence":"disclosed","replacement":{"allowed":false},"submissionBaseline":{"isolationPolicy":"oci-container"}},"protocol":"https://spec.jinn.network/protocols/benchmarking/v1","replicates":3,"venue":{"kind":"self-run"}}