{
  "development_only": true,
  "qualificationGranted": false,
  "unseenComparisonBankInspected": false,
  "frozenPublishedV13Changed": false,
  "initial_runner_sha256": "b24613e249e685e2c2cc6cae9f315148d97c6bfef2c013daa19877128fd166c9",
  "final_runner_sha256": "00d0f68dfa240c7ce937be94ade50172c5051cb243ad50b2740e1b292ea5f65c",
  "offline_profile_sha256": "e5462ef1daae3785012cb8e4fa05b89105bf4ac4d98be9580a8c357365a4f54f",
  "grader_and_inference_guard_functions_unchanged": true,
  "unchanged_function_ast_sha256": {
    "grade": "26c617eb53af3aa94398025211ac6ae868fcbc16127e7a3713666079c4a8ac8b",
    "summary": "8cc75a6b7039f06734e2866cab117ca673277d21980b31b92733d43e47f0bcd4",
    "value_matches": "254cfee1b9a69bfa94f4b6c601c987d58741920df36a596e66011978177337d2",
    "missing_information": "0cb9bae701d15afff51be6696afd7ee85ed296bf07d7e8a783f486696e6935a9",
    "normalize_absence": "ef6912c7106638b8fc536da6a7739fd01f48a79e4a00c42d8c9cbb78395406ec",
    "topic_absence": "bf5d4c792cc5da2564e0a1314c43f7045b04ea8b8726d739fd04e5a87c53351f",
    "documentary_absence_clauses": "3553d9d4802de1ee71f5068160f58389dbdce39703bb86e787b9027d82b66503",
    "ask": "05eae8d73f16afa2226ac86b0eff6c2ac7bdb0e5b0fa4a9bd13fcbdfc9ddb9e2"
  },
  "changes": [
    "Question clauses outrank internal field labels in passage retrieval.",
    "Lexical inflection families cover measurement and finance words without answer facts.",
    "Document identity remains scoped across report/review genre naming.",
    "Distinctive document-name tokens prioritize matching source pages.",
    "Explicit qualifier-bearing passages are preferred when present, allowing approved/signed status synonyms.",
    "Qualifiers stay with the corresponding subquestion rather than spilling into another field."
  ],
  "pure_tests": {
    "exit_code": 0,
    "log_sha256": "c6b9b1f09add0403cc07e90bfdf272129b1ddfa2f8f7675ea8218862f2516371"
  },
  "retained_files": [
    {
      "path": "kingy-repair-1.jsonl",
      "sha256": "2914f1ae6b9fc20ead713bc66a3d24acf43112a3f483a9dbe5779241f3384abd",
      "attempts": 6,
      "passes": 0,
      "failures": 6,
      "model_request_count": 12,
      "answer_seconds": {
        "count": 6,
        "min": 2.9077799579827115,
        "median": 3.7115721455047606,
        "max": 4.49531699999352
      },
      "summary": {
        "runner_version": "kingy-document-runner/1.4.0",
        "questions_sha256": "1abe5e67e1242a335e0493e5c72c8ba4518d22cb09e29c492f72f890858c1e12",
        "references_sha256": "c73ec4c818e4f6aa154a4febf610e2c1fdee1d0edff286013958fd45f1fdd12a",
        "runner_sha256": "b24613e249e685e2c2cc6cae9f315148d97c6bfef2c013daa19877128fd166c9",
        "source_changed_during_run": false,
        "system_prompt_sha256": "cb330d9b009483d2ed91469195ceff86b231fba66feee14ffbdf344153e1ec3a",
        "parse_seconds": 0.039185707981232554,
        "index_seconds": 0.001612167019629851,
        "split": "development",
        "repetitions": [
          {
            "repetition": 1,
            "attempts": 2,
            "answerable_count": 2,
            "answerable_passes": 0,
            "unanswerable_count": 0,
            "correct_abstentions": 0,
            "citation_count": 2,
            "valid_citation_count": 1,
            "answer_correctness": 0.0,
            "citation_precision": 0.5,
            "abstention_accuracy": null,
            "qualified": false,
            "by_category": {
              "cross_document": {
                "attempts": 1,
                "passed": 0
              },
              "direct": {
                "attempts": 1,
                "passed": 0
              }
            },
            "model_calls": 2,
            "model_request_count": 4,
            "guard_only_answers": 0
          },
          {
            "repetition": 2,
            "attempts": 2,
            "answerable_count": 2,
            "answerable_passes": 0,
            "unanswerable_count": 0,
            "correct_abstentions": 0,
            "citation_count": 2,
            "valid_citation_count": 1,
            "answer_correctness": 0.0,
            "citation_precision": 0.5,
            "abstention_accuracy": null,
            "qualified": false,
            "by_category": {
              "cross_document": {
                "attempts": 1,
                "passed": 0
              },
              "direct": {
                "attempts": 1,
                "passed": 0
              }
            },
            "model_calls": 2,
            "model_request_count": 4,
            "guard_only_answers": 0
          },
          {
            "repetition": 3,
            "attempts": 2,
            "answerable_count": 2,
            "answerable_passes": 0,
            "unanswerable_count": 0,
            "correct_abstentions": 0,
            "citation_count": 2,
            "valid_citation_count": 1,
            "answer_correctness": 0.0,
            "citation_precision": 0.5,
            "abstention_accuracy": null,
            "qualified": false,
            "by_category": {
              "cross_document": {
                "attempts": 1,
                "passed": 0
              },
              "direct": {
                "attempts": 1,
                "passed": 0
              }
            },
            "model_calls": 2,
            "model_request_count": 4,
            "guard_only_answers": 0
          }
        ],
        "complete_planned_evaluation": false,
        "qualified": false
      }
    },
    {
      "path": "kingy-repair-2.jsonl",
      "sha256": "d81a67bf20c11681f4ac524fd8c77357a2b648a961cb11ba31649f236a9268a1",
      "attempts": 6,
      "passes": 6,
      "failures": 0,
      "model_request_count": 12,
      "answer_seconds": {
        "count": 6,
        "min": 2.6959760840109084,
        "median": 4.098264083499089,
        "max": 4.593273417034652
      },
      "summary": {
        "runner_version": "kingy-document-runner/1.4.0",
        "questions_sha256": "1abe5e67e1242a335e0493e5c72c8ba4518d22cb09e29c492f72f890858c1e12",
        "references_sha256": "c73ec4c818e4f6aa154a4febf610e2c1fdee1d0edff286013958fd45f1fdd12a",
        "runner_sha256": "c8a7930c185b16567ceefec1049c912b0d2c8c2954d798ad22724e15296726e1",
        "source_changed_during_run": false,
        "system_prompt_sha256": "cb330d9b009483d2ed91469195ceff86b231fba66feee14ffbdf344153e1ec3a",
        "parse_seconds": 0.04120470900670625,
        "index_seconds": 0.001713167002890259,
        "split": "development",
        "repetitions": [
          {
            "repetition": 1,
            "attempts": 2,
            "answerable_count": 2,
            "answerable_passes": 2,
            "unanswerable_count": 0,
            "correct_abstentions": 0,
            "citation_count": 3,
            "valid_citation_count": 3,
            "answer_correctness": 1.0,
            "citation_precision": 1.0,
            "abstention_accuracy": null,
            "qualified": false,
            "by_category": {
              "cross_document": {
                "attempts": 1,
                "passed": 1
              },
              "direct": {
                "attempts": 1,
                "passed": 1
              }
            },
            "model_calls": 2,
            "model_request_count": 4,
            "guard_only_answers": 0
          },
          {
            "repetition": 2,
            "attempts": 2,
            "answerable_count": 2,
            "answerable_passes": 2,
            "unanswerable_count": 0,
            "correct_abstentions": 0,
            "citation_count": 3,
            "valid_citation_count": 3,
            "answer_correctness": 1.0,
            "citation_precision": 1.0,
            "abstention_accuracy": null,
            "qualified": false,
            "by_category": {
              "cross_document": {
                "attempts": 1,
                "passed": 1
              },
              "direct": {
                "attempts": 1,
                "passed": 1
              }
            },
            "model_calls": 2,
            "model_request_count": 4,
            "guard_only_answers": 0
          },
          {
            "repetition": 3,
            "attempts": 2,
            "answerable_count": 2,
            "answerable_passes": 2,
            "unanswerable_count": 0,
            "correct_abstentions": 0,
            "citation_count": 3,
            "valid_citation_count": 3,
            "answer_correctness": 1.0,
            "citation_precision": 1.0,
            "abstention_accuracy": null,
            "qualified": false,
            "by_category": {
              "cross_document": {
                "attempts": 1,
                "passed": 1
              },
              "direct": {
                "attempts": 1,
                "passed": 1
              }
            },
            "model_calls": 2,
            "model_request_count": 4,
            "guard_only_answers": 0
          }
        ],
        "complete_planned_evaluation": false,
        "qualified": false
      }
    },
    {
      "path": "kingy-repair-3.jsonl",
      "sha256": "0a4f915bbcb9c9377525ebffdcf039588024baf4fa5550982fe311666f04d349",
      "attempts": 6,
      "passes": 6,
      "failures": 0,
      "model_request_count": 12,
      "answer_seconds": {
        "count": 6,
        "min": 2.9620087079820223,
        "median": 3.7256092919851653,
        "max": 5.28013720799936
      },
      "summary": {
        "runner_version": "kingy-document-runner/1.4.0",
        "questions_sha256": "1abe5e67e1242a335e0493e5c72c8ba4518d22cb09e29c492f72f890858c1e12",
        "references_sha256": "c73ec4c818e4f6aa154a4febf610e2c1fdee1d0edff286013958fd45f1fdd12a",
        "runner_sha256": "00d0f68dfa240c7ce937be94ade50172c5051cb243ad50b2740e1b292ea5f65c",
        "source_changed_during_run": false,
        "system_prompt_sha256": "cb330d9b009483d2ed91469195ceff86b231fba66feee14ffbdf344153e1ec3a",
        "parse_seconds": 0.043701082991901785,
        "index_seconds": 0.0017875409976113588,
        "split": "development",
        "repetitions": [
          {
            "repetition": 1,
            "attempts": 2,
            "answerable_count": 2,
            "answerable_passes": 2,
            "unanswerable_count": 0,
            "correct_abstentions": 0,
            "citation_count": 3,
            "valid_citation_count": 3,
            "answer_correctness": 1.0,
            "citation_precision": 1.0,
            "abstention_accuracy": null,
            "qualified": false,
            "by_category": {
              "cross_document": {
                "attempts": 1,
                "passed": 1
              },
              "direct": {
                "attempts": 1,
                "passed": 1
              }
            },
            "model_calls": 2,
            "model_request_count": 4,
            "guard_only_answers": 0
          },
          {
            "repetition": 2,
            "attempts": 2,
            "answerable_count": 2,
            "answerable_passes": 2,
            "unanswerable_count": 0,
            "correct_abstentions": 0,
            "citation_count": 3,
            "valid_citation_count": 3,
            "answer_correctness": 1.0,
            "citation_precision": 1.0,
            "abstention_accuracy": null,
            "qualified": false,
            "by_category": {
              "cross_document": {
                "attempts": 1,
                "passed": 1
              },
              "direct": {
                "attempts": 1,
                "passed": 1
              }
            },
            "model_calls": 2,
            "model_request_count": 4,
            "guard_only_answers": 0
          },
          {
            "repetition": 3,
            "attempts": 2,
            "answerable_count": 2,
            "answerable_passes": 2,
            "unanswerable_count": 0,
            "correct_abstentions": 0,
            "citation_count": 3,
            "valid_citation_count": 3,
            "answer_correctness": 1.0,
            "citation_precision": 1.0,
            "abstention_accuracy": null,
            "qualified": false,
            "by_category": {
              "cross_document": {
                "attempts": 1,
                "passed": 1
              },
              "direct": {
                "attempts": 1,
                "passed": 1
              }
            },
            "model_calls": 2,
            "model_request_count": 4,
            "guard_only_answers": 0
          }
        ],
        "complete_planned_evaluation": false,
        "qualified": false
      }
    },
    {
      "path": "regression-17-2.jsonl",
      "sha256": "a60a9bcfd4af51d8ac11efdebb15da362f077a52e993da9a6e87592064f0c222",
      "attempts": 17,
      "passes": 17,
      "failures": 0,
      "model_request_count": 34,
      "answer_seconds": {
        "count": 17,
        "min": 2.418795457982924,
        "median": 2.9926566669892054,
        "max": 4.760433417017339
      },
      "summary": {
        "runner_version": "kingy-document-runner/1.4.0",
        "questions_sha256": "a69c2f7f8ba63c939d5b226a6a392975af67f5f11d433eecf571ca64a230b469",
        "references_sha256": "590bbcfb17ec5d3f61323fc40a2a6439d266e62e69a88a489bfc6f008791dccb",
        "runner_sha256": "00d0f68dfa240c7ce937be94ade50172c5051cb243ad50b2740e1b292ea5f65c",
        "source_changed_during_run": false,
        "system_prompt_sha256": "cb330d9b009483d2ed91469195ceff86b231fba66feee14ffbdf344153e1ec3a",
        "parse_seconds": 0.03935916698537767,
        "index_seconds": 0.0015818749961908907,
        "split": "development",
        "repetitions": [
          {
            "repetition": 1,
            "attempts": 17,
            "answerable_count": 13,
            "answerable_passes": 13,
            "unanswerable_count": 4,
            "correct_abstentions": 4,
            "citation_count": 20,
            "valid_citation_count": 20,
            "answer_correctness": 1.0,
            "citation_precision": 1.0,
            "abstention_accuracy": 1.0,
            "qualified": true,
            "by_category": {
              "cross_document": {
                "attempts": 7,
                "passed": 7
              },
              "direct": {
                "attempts": 6,
                "passed": 6
              },
              "unanswerable": {
                "attempts": 4,
                "passed": 4
              }
            },
            "model_calls": 17,
            "model_request_count": 34,
            "guard_only_answers": 0
          }
        ],
        "complete_planned_evaluation": false,
        "qualified": false
      }
    },
    {
      "path": "regression-17.jsonl",
      "sha256": "5e0eff58b18b7ecc3f33a101e4bf0ff3b6bb25ac8e6bd73e8fd2d6b6df5fff36",
      "attempts": 17,
      "passes": 15,
      "failures": 2,
      "model_request_count": 34,
      "answer_seconds": {
        "count": 17,
        "min": 2.2564609999826644,
        "median": 2.9022894999943674,
        "max": 3.903823333006585
      },
      "summary": {
        "runner_version": "kingy-document-runner/1.4.0",
        "questions_sha256": "a69c2f7f8ba63c939d5b226a6a392975af67f5f11d433eecf571ca64a230b469",
        "references_sha256": "590bbcfb17ec5d3f61323fc40a2a6439d266e62e69a88a489bfc6f008791dccb",
        "runner_sha256": "c8a7930c185b16567ceefec1049c912b0d2c8c2954d798ad22724e15296726e1",
        "source_changed_during_run": false,
        "system_prompt_sha256": "cb330d9b009483d2ed91469195ceff86b231fba66feee14ffbdf344153e1ec3a",
        "parse_seconds": 0.0353045419906266,
        "index_seconds": 0.0014627919881604612,
        "split": "development",
        "repetitions": [
          {
            "repetition": 1,
            "attempts": 17,
            "answerable_count": 13,
            "answerable_passes": 11,
            "unanswerable_count": 4,
            "correct_abstentions": 4,
            "citation_count": 18,
            "valid_citation_count": 17,
            "answer_correctness": 0.8461538461538461,
            "citation_precision": 0.9444444444444444,
            "abstention_accuracy": 1.0,
            "qualified": false,
            "by_category": {
              "cross_document": {
                "attempts": 7,
                "passed": 5
              },
              "direct": {
                "attempts": 6,
                "passed": 6
              },
              "unanswerable": {
                "attempts": 4,
                "passed": 4
              }
            },
            "model_calls": 17,
            "model_request_count": 34,
            "guard_only_answers": 0
          }
        ],
        "complete_planned_evaluation": false,
        "qualified": false
      }
    },
    {
      "path": "regression-9.jsonl",
      "sha256": "8d9d3f573363ad70e4d58ed4a7a9ec6f6a800afb80a8390a5c49079b31cc332c",
      "attempts": 9,
      "passes": 9,
      "failures": 0,
      "model_request_count": 18,
      "answer_seconds": {
        "count": 9,
        "min": 2.7322549159871414,
        "median": 2.7946781659848057,
        "max": 3.2021582489833236
      },
      "summary": {
        "runner_version": "kingy-document-runner/1.4.0",
        "questions_sha256": "1a3537084889d8cb7f61f2b40a64f1cbf4b3d62a283b2b81ce4214900d84c103",
        "references_sha256": "3ec6bd67d8fb393ef269cdfa51ef1f949e21de6ed82de3aba948eb9c186c593e",
        "runner_sha256": "00d0f68dfa240c7ce937be94ade50172c5051cb243ad50b2740e1b292ea5f65c",
        "source_changed_during_run": false,
        "system_prompt_sha256": "cb330d9b009483d2ed91469195ceff86b231fba66feee14ffbdf344153e1ec3a",
        "parse_seconds": 0.042581790999975055,
        "index_seconds": 0.0017281250038649887,
        "split": "development",
        "repetitions": [
          {
            "repetition": 1,
            "attempts": 9,
            "answerable_count": 0,
            "answerable_passes": 0,
            "unanswerable_count": 9,
            "correct_abstentions": 9,
            "citation_count": 0,
            "valid_citation_count": 0,
            "answer_correctness": null,
            "citation_precision": null,
            "abstention_accuracy": 1.0,
            "qualified": false,
            "by_category": {
              "unanswerable": {
                "attempts": 9,
                "passed": 9
              }
            },
            "model_calls": 9,
            "model_request_count": 18,
            "guard_only_answers": 0
          }
        ],
        "complete_planned_evaluation": false,
        "qualified": false
      }
    },
    {
      "path": "source-controls.jsonl",
      "sha256": "6814efd59fc010a24823204140c4147a8347b08c020e9a6535ed502cb3c8e0b4",
      "attempts": 30,
      "passes": 30,
      "failures": 0,
      "model_request_count": 60,
      "answer_seconds": {
        "count": 30,
        "min": 1.674753208993934,
        "median": 1.7648069164861226,
        "max": 1.9064500829845201
      },
      "summary": {
        "runner_sha256": "00d0f68dfa240c7ce937be94ade50172c5051cb243ad50b2740e1b292ea5f65c",
        "source_changed_during_run": false,
        "development_only": true,
        "qualificationGranted": false,
        "attempts": 30,
        "model_request_count": 60,
        "passed": true,
        "failures": []
      }
    }
  ],
  "all_retained_attempts": 91,
  "all_retained_model_request_count": 182,
  "pre_model_invocation_errors": [
    {
      "commandSplit": "heldout",
      "bank": "consumed regression-v12 development split",
      "error": "Selected question split is empty",
      "model_requests": 0,
      "resolution": "Re-ran with the bank existing development split; original question/reference files unchanged."
    }
  ],
  "limitations": [
    "Consumed regressions and synthetic controls cannot qualify a release or prove comparative accuracy.",
    "No model server was started or stopped by this repair task.",
    "No extra model self-correction call was added; inference remains source selection plus semantic sufficiency checking."
  ]
}
