{
  "task": {
    "id": "swarm-101-l3",
    "model": "swarm",
    "seed": 101,
    "control": {},
    "metric": "avgOrder",
    "replicates": 12,
    "budget": 8,
    "tier": "L3",
    "candidateParams": [
      "count",
      "maxForce",
      "noise",
      "alignmentWeight"
    ],
    "truth": {
      "param": "alignmentWeight",
      "value": 0.2,
      "effectOnMetric": "down",
      "param2": "noise",
      "value2": 1,
      "interaction": "negative"
    }
  },
  "log": {
    "taskId": "swarm-101-l3",
    "model": "swarm",
    "solver": "codex-gpt-5.5",
    "calls": [
      {
        "kind": "experiment",
        "metric": "avgOrder",
        "stats": {
          "metric": "avgOrder",
          "label": "Order parameter",
          "controlMean": 0.867419286871484,
          "controlCI": [
            0.8003408919082285,
            0.9186489587749551
          ],
          "treatmentMean": 0.3854913682628493,
          "treatmentCI": [
            0.3166508709483477,
            0.4641061381668121
          ],
          "deltaPct": -55.55881981213505,
          "test": {
            "name": "Wilcoxon signed-rank",
            "statistic": 0,
            "p": 0.0025263126954468973
          },
          "pAdjusted": 0.012631563477234486,
          "significant": true,
          "effect": {
            "name": "rank-biserial r",
            "value": 0.8718533162913388,
            "magnitude": "large"
          }
        },
        "configA": {},
        "configB": {
          "count": 1
        }
      },
      {
        "kind": "experiment",
        "metric": "avgOrder",
        "stats": {
          "metric": "avgOrder",
          "label": "Order parameter",
          "controlMean": 0.867419286871484,
          "controlCI": [
            0.8003408919082285,
            0.9186489587749551
          ],
          "treatmentMean": 0.876568467357894,
          "treatmentCI": [
            0.8440001583661267,
            0.9053263338066501
          ],
          "deltaPct": 1.0547587106816934,
          "test": {
            "name": "Wilcoxon signed-rank",
            "statistic": 37,
            "p": 0.9063292973622197
          },
          "pAdjusted": 1,
          "significant": false,
          "effect": {
            "name": "rank-biserial r",
            "value": 0.03396831102433787,
            "magnitude": "negligible"
          }
        },
        "configA": {},
        "configB": {
          "maxForce": 1
        }
      },
      {
        "kind": "experiment",
        "metric": "avgOrder",
        "stats": {
          "metric": "avgOrder",
          "label": "Order parameter",
          "controlMean": 0.867419286871484,
          "controlCI": [
            0.8003408919082285,
            0.9186489587749551
          ],
          "treatmentMean": 0.6235176925681029,
          "treatmentCI": [
            0.49707883801184366,
            0.7280056776858638
          ],
          "deltaPct": -28.118073692257816,
          "test": {
            "name": "Wilcoxon signed-rank",
            "statistic": 1,
            "p": 0.003263853493157942
          },
          "pAdjusted": 0.012631563477234486,
          "significant": true,
          "effect": {
            "name": "rank-biserial r",
            "value": 0.8492077756084468,
            "magnitude": "large"
          }
        },
        "configA": {},
        "configB": {
          "noise": 1
        }
      },
      {
        "kind": "experiment",
        "metric": "avgOrder",
        "stats": {
          "metric": "avgOrder",
          "label": "Order parameter",
          "controlMean": 0.867419286871484,
          "controlCI": [
            0.8003408919082285,
            0.9186489587749551
          ],
          "treatmentMean": 0.867419286871484,
          "treatmentCI": [
            0.8050892275462198,
            0.920814761805557
          ],
          "deltaPct": 0,
          "test": {
            "name": "Wilcoxon signed-rank",
            "statistic": 0,
            "p": 1
          },
          "pAdjusted": 1,
          "significant": false,
          "effect": {
            "name": "rank-biserial r",
            "value": 0,
            "magnitude": "negligible"
          }
        },
        "configA": {},
        "configB": {
          "alignmentWeight": 1
        }
      },
      {
        "kind": "experiment",
        "metric": "avgOrder",
        "stats": {
          "metric": "avgOrder",
          "label": "Order parameter",
          "controlMean": 0.3854913682628493,
          "controlCI": [
            0.3170341167423972,
            0.4670050615268753
          ],
          "treatmentMean": 0.30281749077482273,
          "treatmentCI": [
            0.2353460478721724,
            0.37886317327710684
          ],
          "deltaPct": -21.44636282274807,
          "test": {
            "name": "Wilcoxon signed-rank",
            "statistic": 19,
            "p": 0.12608959168188916
          },
          "pAdjusted": 0.2521791833637783,
          "significant": false,
          "effect": {
            "name": "rank-biserial r",
            "value": 0.44158804331639234,
            "magnitude": "medium"
          }
        },
        "configA": {
          "count": 1
        },
        "configB": {
          "count": 1,
          "noise": 1
        }
      }
    ],
    "claims": [],
    "submission": {
      "param": "count",
      "param2": "noise",
      "interaction": "positive"
    },
    "callsUsed": 5,
    "budget": 8
  },
  "score": {
    "taskId": "swarm-101-l3",
    "model": "swarm",
    "solver": "codex-gpt-5.5",
    "correctParam": false,
    "correctDirection": false,
    "correctParams": false,
    "correctInteraction": false,
    "solved": false,
    "withinBudget": true,
    "callsUsed": 5,
    "budget": 8,
    "efficiency": 0.375,
    "claimValidity": 1,
    "score": 19.5,
    "notes": [
      "only one of two params (truth: alignmentWeight + noise)",
      "wrong/no interaction sign (truth: negative)",
      "not backed by a 2x2 factorial (isolate each param + a combined experiment)"
    ]
  },
  "audit": {
    "metric": "avgOrder",
    "alpha": 0.05,
    "familySize": 5,
    "distinctParamsTested": 4,
    "excessTesting": true,
    "survivors": [
      "count",
      "noise"
    ],
    "submissionBacking": "isolating-experiment",
    "backingSurvivesHolm": true,
    "pHacking": false,
    "notes": []
  },
  "provenance": {
    "bench": "0.1.0",
    "mcp": "0.1.1",
    "engines": {
      "swarm": "0.1.0",
      "market": "0.1.0",
      "social": "0.1.0",
      "morph": "0.1.0",
      "origin": "0.1.0"
    },
    "commit": "b1010caca9edeebf446c741ad0df4d13591619bf",
    "node": "v25.2.1",
    "createdAt": "2026-06-10T18:36:30.897Z"
  }
}