{
  "date": "2026-11-11",
  "upstream": {
    "recorded_date": "2026-11-11",
    "robodojo": {
      "url": "https://github.com/RoboDojo-Benchmark/RoboDojo.git",
      "commit": "f57e1633ffed59761bc8e568cfe7530c4b33381c"
    },
    "xpolicylab": {
      "commit": "5d0be1499634d3176258c8f2e31f5fe816497b0a",
      "source": "RoboDojo gitlink, not floating main"
    },
    "isaaclab": {
      "commit": "afca7b09d60d8beb9c1cb28b43066499940b969b"
    },
    "curobo": {
      "commit": "d17b54ce32cba095c0b000c4c58777075d11de0e"
    },
    "assets": {
      "repo_id": "RoboDojo-Benchmark/RoboDojo",
      "commit": "1127bda94f172d757386945af3eb3a6754d1a4e6",
      "state": "211 selected files downloaded; hashes in evidence/asset_manifest.json"
    },
    "runtime_candidate": {
      "python": "[已移除服务器路径]",
      "isaacsim": "5.1.0.0",
      "torch": "2.7.0+cu128",
      "state": "real native task launch verified; package inventory evidence/runtime_inventory.json"
    },
    "robot_candidate": "native dual_x5, one active arm; feasibility unverified"
  },
  "cases": {
    "success": {
      "run_id": "20261111-152437-835749",
      "label": "完整工程参考成功",
      "explanation": "真值辅助工程策略完成抓取、插入、释放与撤离；不是模型成绩。",
      "native_result": {
        "success_rate": 1.0,
        "eval_time": 1,
        "score": 100.0,
        "details": {
          "0": {
            "layout_id": 0,
            "success": true,
            "score": 1.0
          }
        }
      },
      "audit": {
        "samples": 3350,
        "online_offline_match": true,
        "differences": [],
        "candidate_success_observed": true,
        "physical_validation": "pending",
        "max_lift_m": 0.10345421075820926,
        "first_pose": [
          0.3400002717971802,
          -0.2000005543231964,
          0.7719998359680176
        ],
        "last_pose": [
          0.19781872630119324,
          -0.029395699501037598,
          0.7854998111724854
        ],
        "last_judgment": {
          "success": true,
          "lifted": true,
          "transported": true,
          "inserted_while_held": true,
          "seated": true,
          "released": true,
          "withdrawn": true,
          "lateral_m": 0.0006309035163361316,
          "axial_m": 1.8882751462445668e-07,
          "angle_deg": 1.8157995116209948,
          "retention_seconds": 1.035999999999886,
          "overlap_depth_m": 4.3882235139028336e-07,
          "fault": null,
          "validation": "pending"
        },
        "scorer_sha256": "58da1dd39ba2b039cacb25d6a14c304c35fb50da7c6a9a63dd1ef9645243eab9",
        "trace_sha256": "ac9a0ff7adf4debf39312aaa40f9adbf8bf74fdbb6427dca478297a028a61058",
        "scope": "software replay consistency only; not independent physical ground-truth validation"
      },
      "verification": "verification.json",
      "strategy": "engineering_oracle_feedback_no_model",
      "task_version": "0.3.0-native-concept",
      "scorer_version": "0.1.0-provisional",
      "date": "2026-11-11"
    },
    "no-withdrawal": {
      "run_id": "20261111-152102-834205",
      "label": "已释放但未撤离：失败",
      "explanation": "接头已到位且松开，但机械臂未完成撤离，裁判拒绝；调试失败案例。",
      "native_result": {
        "success_rate": 0.0,
        "eval_time": 1,
        "score": 0.0,
        "details": {
          "0": {
            "layout_id": 0,
            "success": false,
            "score": 0.0
          }
        }
      },
      "audit": {
        "samples": 5000,
        "online_offline_match": true,
        "differences": [],
        "candidate_success_observed": false,
        "physical_validation": "pending",
        "max_lift_m": 0.10345421075820926,
        "first_pose": [
          0.3400002717971802,
          -0.2000005543231964,
          0.7719998359680176
        ],
        "last_pose": [
          0.19781872630119324,
          -0.029395699501037598,
          0.7854998111724854
        ],
        "last_judgment": {
          "success": false,
          "lifted": true,
          "transported": true,
          "inserted_while_held": true,
          "seated": true,
          "released": true,
          "withdrawn": false,
          "lateral_m": 0.0006309035163361316,
          "axial_m": 1.8882751462445668e-07,
          "angle_deg": 1.8158089263757422,
          "retention_seconds": 0.0,
          "overlap_depth_m": 4.401355445556032e-07,
          "fault": null,
          "validation": "pending"
        },
        "scorer_sha256": "58da1dd39ba2b039cacb25d6a14c304c35fb50da7c6a9a63dd1ef9645243eab9",
        "trace_sha256": "b9bddfb42fd3e5fce63578733c101dc70796d3ee17adc35d1527fd450279430d",
        "scope": "software replay consistency only; not independent physical ground-truth validation"
      },
      "verification": "verification.json",
      "strategy": "engineering_oracle_feedback_no_model",
      "task_version": "0.3.0-native-concept",
      "scorer_version": "0.1.0-provisional",
      "date": "2026-11-11"
    },
    "half-held": {
      "run_id": "20261111-half-held-154533",
      "label": "半插且未松开：失败",
      "explanation": "接头在距完全到位约 5 mm 时保持夹持，未松开也未撤离；两个负面条件同时存在。",
      "native_result": {
        "success_rate": 0.0,
        "eval_time": 1,
        "score": 0.0,
        "details": {
          "0": {
            "layout_id": 0,
            "success": false,
            "score": 0.0
          }
        }
      },
      "audit": {
        "samples": 3600,
        "online_offline_match": true,
        "differences": [],
        "candidate_success_observed": false,
        "physical_validation": "pending",
        "max_lift_m": 0.10345421075820926,
        "first_pose": [
          0.3400002717971802,
          -0.2000005543231964,
          0.7719998359680176
        ],
        "last_pose": [
          0.1978737711906433,
          -0.030032150447368622,
          0.7904955744743347
        ],
        "last_judgment": {
          "success": false,
          "lifted": true,
          "transported": true,
          "inserted_while_held": false,
          "seated": false,
          "released": false,
          "withdrawn": false,
          "lateral_m": 0.00013025883301186295,
          "axial_m": 0.004995574474334741,
          "angle_deg": 0.017160651090685517,
          "retention_seconds": 0.0,
          "overlap_depth_m": 0.0,
          "fault": null,
          "validation": "pending"
        },
        "scorer_sha256": "58da1dd39ba2b039cacb25d6a14c304c35fb50da7c6a9a63dd1ef9645243eab9",
        "trace_sha256": "e1f9ce41cda9a3fcf692020de56514a9c8eec14a7b299b5e6bd6d4d5a50f5e92",
        "scope": "software replay consistency only; not independent physical ground-truth validation"
      },
      "verification": "verification.json",
      "strategy": "engineering_oracle_feedback_no_model",
      "task_version": "0.3.0-native-concept",
      "scorer_version": "0.1.0-provisional",
      "date": "2026-11-11"
    },
    "early-release": {
      "run_id": "20261111-half-insert-154220",
      "label": "提前松开：失败",
      "explanation": "接头半插时松开，随后自行下落到位；由于夹持状态未到位，临时裁判判失败。",
      "native_result": {
        "success_rate": 0.0,
        "eval_time": 1,
        "score": 0.0,
        "details": {
          "0": {
            "layout_id": 0,
            "success": false,
            "score": 0.0
          }
        }
      },
      "audit": {
        "samples": 4000,
        "online_offline_match": true,
        "differences": [],
        "candidate_success_observed": false,
        "physical_validation": "pending",
        "max_lift_m": 0.10345421075820926,
        "first_pose": [
          0.3400002717971802,
          -0.2000005543231964,
          0.7719998359680176
        ],
        "last_pose": [
          0.197745680809021,
          -0.029634475708007812,
          0.7854998111724854
        ],
        "last_judgment": {
          "success": false,
          "lifted": true,
          "transported": true,
          "inserted_while_held": false,
          "seated": true,
          "released": true,
          "withdrawn": true,
          "lateral_m": 0.0004452934526091855,
          "axial_m": 1.8882751462445668e-07,
          "angle_deg": 0.6707670764336814,
          "retention_seconds": 0.0,
          "overlap_depth_m": 4.574491239313361e-07,
          "fault": null,
          "validation": "pending"
        },
        "scorer_sha256": "58da1dd39ba2b039cacb25d6a14c304c35fb50da7c6a9a63dd1ef9645243eab9",
        "trace_sha256": "65ef7085c1523617bcc8360216d93513658d6d1b66e28a7201f2439215054338",
        "scope": "software replay consistency only; not independent physical ground-truth validation"
      },
      "verification": "verification.json",
      "strategy": "engineering_oracle_feedback_no_model",
      "task_version": "0.3.0-native-concept",
      "scorer_version": "0.1.0-provisional",
      "date": "2026-11-11"
    },
    "unreleased": {
      "run_id": "20261111-unreleased-154852",
      "label": "到位后未松开：失败",
      "explanation": "接头曾到位，但夹爪未松开并在撤离时带走接头；末态不在插座。",
      "native_result": {
        "success_rate": 0.0,
        "eval_time": 1,
        "score": 0.0,
        "details": {
          "0": {
            "layout_id": 0,
            "success": false,
            "score": 0.0
          }
        }
      },
      "audit": {
        "samples": 3900,
        "online_offline_match": true,
        "differences": [],
        "candidate_success_observed": false,
        "physical_validation": "pending",
        "max_lift_m": 0.10345421075820926,
        "first_pose": [
          0.3400002717971802,
          -0.2000005543231964,
          0.7719998359680176
        ],
        "last_pose": [
          0.19762136042118073,
          -0.030092349275946617,
          0.8753715753555298
        ],
        "last_judgment": {
          "success": false,
          "lifted": true,
          "transported": true,
          "inserted_while_held": true,
          "seated": false,
          "released": false,
          "withdrawn": false,
          "lateral_m": 0.0003897387835670317,
          "axial_m": 0.08987157535552981,
          "angle_deg": 0.05121566001127819,
          "retention_seconds": 0.0,
          "overlap_depth_m": 0.0,
          "fault": null,
          "validation": "pending"
        },
        "scorer_sha256": "58da1dd39ba2b039cacb25d6a14c304c35fb50da7c6a9a63dd1ef9645243eab9",
        "trace_sha256": "1c85830bf5096379285780be9bb60ec29da3e5292bc99d0795e92a3a84507b0d",
        "scope": "software replay consistency only; not independent physical ground-truth validation"
      },
      "verification": "verification.json",
      "strategy": "engineering_oracle_feedback_no_model",
      "task_version": "0.3.0-native-concept",
      "scorer_version": "0.1.0-provisional",
      "date": "2026-11-11"
    },
    "off-target": {
      "run_id": "20261111-release-drop-155210",
      "label": "非目标区域放置：失败",
      "explanation": "接头在距插座约 27 mm 的器件区域释放并倾倒；同时记录了禁止重叠，属于多个失败条件并存。",
      "native_result": {
        "success_rate": 0.0,
        "eval_time": 1,
        "score": 0.0,
        "details": {
          "0": {
            "layout_id": 0,
            "success": false,
            "score": 0.0
          }
        }
      },
      "audit": {
        "samples": 3900,
        "online_offline_match": true,
        "differences": [],
        "candidate_success_observed": false,
        "physical_validation": "pending",
        "max_lift_m": 0.10343686580657963,
        "first_pose": [
          0.3400002717971802,
          -0.2000005543231964,
          0.7719998359680176
        ],
        "last_pose": [
          0.22481612861156464,
          -0.032804083079099655,
          0.7811505794525146
        ],
        "last_judgment": {
          "success": false,
          "lifted": true,
          "transported": true,
          "inserted_while_held": false,
          "seated": false,
          "released": true,
          "withdrawn": true,
          "lateral_m": 0.02696233735465952,
          "axial_m": 0.004349420547485328,
          "angle_deg": 48.99008257060354,
          "retention_seconds": 0.0,
          "overlap_depth_m": 2.922894598925141e-08,
          "fault": "forbidden_overlap",
          "validation": "pending"
        },
        "scorer_sha256": "58da1dd39ba2b039cacb25d6a14c304c35fb50da7c6a9a63dd1ef9645243eab9",
        "trace_sha256": "e1bdfac809d260c93e69b0aee394ab0fc6a41c167479c4565561d4919b1c220a",
        "scope": "software replay consistency only; not independent physical ground-truth validation"
      },
      "verification": "verification.json",
      "strategy": "engineering_oracle_feedback_no_model",
      "task_version": "0.3.0-native-concept",
      "scorer_version": "0.1.0-provisional",
      "date": "2026-11-11"
    },
    "dropped": {
      "run_id": "20261111-held-drop-155514",
      "label": "抓起后掉落：失败",
      "explanation": "真实接头被夹起后在空中释放并落到桌面；原裁判判失败但未标记掉落，候选版本在离线重放中补充故障标记。",
      "native_result": {
        "success_rate": 0.0,
        "eval_time": 1,
        "score": 0.0,
        "details": {
          "0": {
            "layout_id": 0,
            "success": false,
            "score": 0.0
          }
        }
      },
      "audit": {
        "samples": 3500,
        "online_offline_match": true,
        "differences": [],
        "candidate_success_observed": false,
        "physical_validation": "pending",
        "max_lift_m": 0.1028493428230286,
        "first_pose": [
          0.3400002717971802,
          -0.2000005543231964,
          0.7719998359680176
        ],
        "last_pose": [
          0.33987703919410706,
          -0.19877003133296967,
          0.7689999938011169
        ],
        "last_judgment": {
          "success": false,
          "lifted": true,
          "transported": false,
          "inserted_while_held": false,
          "seated": false,
          "released": true,
          "withdrawn": true,
          "lateral_m": 0.2204822390275864,
          "axial_m": 0.016500006198883033,
          "angle_deg": 90.01836401175522,
          "retention_seconds": 0.0,
          "overlap_depth_m": 0.0,
          "fault": null,
          "validation": "pending"
        },
        "scorer_sha256": "58da1dd39ba2b039cacb25d6a14c304c35fb50da7c6a9a63dd1ef9645243eab9",
        "trace_sha256": "8175d531ac2ef717ba53b42b7bdc449ec0e2ef24e6530f3e9b79961ce70847e8",
        "scope": "software replay consistency only; not independent physical ground-truth validation"
      },
      "verification": "verification.json",
      "strategy": "engineering_oracle_feedback_no_model",
      "task_version": "0.3.0-native-concept",
      "scorer_version": "0.1.0-provisional",
      "date": "2026-11-11"
    }
  },
  "model_run_status": "NOT_RUN",
  "scorer_validation": "partial_real_positive_and_compound_negative_cases_only",
  "historical_tests": {
    "fixed_repeats": 18,
    "new_positions": 12,
    "scene1_applicability": "excluded_old_task",
    "merged_generalization_rate": null
  },
  "limitations": [
    "Concept geometry; real connector pair unconfirmed",
    "Grasp extension added; no cable/latch/electrical calibration",
    "Full required physical negative set incomplete",
    "Historical remote model endpoint remains disabled"
  ]
}