{
 "schemaVersion": "blobfish.typed-decision-bench-tasks.v1",
 "version": "0.3.0",
 "observedAt": "2026-09-20T20:01:35Z",
 "systems": [
  "Jev",
  "JevFish",
  "NanoJev",
  "OpenJev",
  "System One Scorer",
  "decider-2b"
 ],
 "tasks": [
  {
   "task": "abt-buy-product-match",
   "family": "retrieval-and-knowledge",
   "use_case": "entity-alignment",
   "vertical": "e-commerce",
   "dataset": "matchbench/Abt-Buy",
   "licence": "CC (variant unconfirmed on the source page) — FLAGGED",
   "provenance": "expert",
   "primitive": "score",
   "items": 200,
   "scores": {
    "decider-2b": {
     "decisionScore": 82.7,
     "accuracy": 0.905,
     "answered": 200,
     "items": 200
    },
    "Jev": {
     "decisionScore": 94.91,
     "accuracy": 0.93,
     "answered": 200,
     "items": 200
    },
    "JevFish": {
     "decisionScore": 87.1,
     "accuracy": 0.895,
     "answered": 200,
     "items": 200
    },
    "NanoJev": {
     "decisionScore": 74.54,
     "accuracy": 0.665,
     "answered": 200,
     "items": 200
    },
    "OpenJev": {
     "decisionScore": 92.91,
     "accuracy": 0.92,
     "answered": 200,
     "items": 200
    },
    "System One Scorer": {
     "decisionScore": 85.37,
     "accuracy": 0.865,
     "answered": 200,
     "items": 200
    }
   }
  },
  {
   "task": "arxiv-archive-subcategory",
   "family": "data-and-operations",
   "use_case": "hierarchical-classification",
   "vertical": "science-publishing",
   "dataset": "gfissore/arxiv-abstracts-2021",
   "licence": "CC0-1.0",
   "provenance": "outcome",
   "primitive": "choice",
   "items": 200,
   "scores": {
    "decider-2b": {
     "decisionScore": 70.09,
     "accuracy": 0.78,
     "answered": 200,
     "items": 200
    },
    "Jev": {
     "decisionScore": 76.85,
     "accuracy": 0.885,
     "answered": 200,
     "items": 200
    },
    "JevFish": {
     "decisionScore": 70.72,
     "accuracy": 0.81,
     "answered": 200,
     "items": 200
    },
    "NanoJev": {
     "decisionScore": 53.67,
     "accuracy": 0.295,
     "answered": 200,
     "items": 200
    },
    "OpenJev": {
     "decisionScore": 65.41,
     "accuracy": 0.865,
     "answered": 200,
     "items": 200
    },
    "System One Scorer": {
     "decisionScore": 70.57,
     "accuracy": 0.815,
     "answered": 200,
     "items": 200
    }
   }
  },
  {
   "task": "bfcl-tool-dispatch",
   "family": "real-time-and-agents",
   "use_case": "agent-routing-and-skill-selection",
   "vertical": "software-engineering",
   "dataset": "minpeter/bfcl-v1-non-live-ast-parsed",
   "licence": "Apache-2.0",
   "provenance": "expert",
   "primitive": "choice",
   "items": 200,
   "scores": {
    "decider-2b": {
     "decisionScore": 92.13,
     "accuracy": 0.89,
     "answered": 200,
     "items": 200
    },
    "Jev": {
     "decisionScore": 97.56,
     "accuracy": 0.97,
     "answered": 200,
     "items": 200
    },
    "JevFish": {
     "decisionScore": 93.01,
     "accuracy": 0.895,
     "answered": 200,
     "items": 200
    },
    "NanoJev": {
     "decisionScore": 61.03,
     "accuracy": 0.4541,
     "answered": 196,
     "items": 200
    },
    "OpenJev": {
     "decisionScore": 96.45,
     "accuracy": 0.96,
     "answered": 200,
     "items": 200
    },
    "System One Scorer": {
     "decisionScore": 83.18,
     "accuracy": 0.79,
     "answered": 200,
     "items": 200
    }
   }
  },
  {
   "task": "casehold-cited-holding",
   "family": "retrieval-and-knowledge",
   "use_case": "citation-verification",
   "vertical": "legal",
   "dataset": "coastalcph/lex_glue",
   "licence": "CC-BY-4.0",
   "provenance": "outcome",
   "primitive": "choice",
   "items": 200,
   "scores": {
    "decider-2b": {
     "decisionScore": 79.98,
     "accuracy": 0.705,
     "answered": 200,
     "items": 200
    },
    "Jev": {
     "decisionScore": 87.03,
     "accuracy": 0.795,
     "answered": 200,
     "items": 200
    },
    "JevFish": {
     "decisionScore": 80.0,
     "accuracy": 0.745,
     "answered": 200,
     "items": 200
    },
    "NanoJev": {
     "decisionScore": 61.18,
     "accuracy": 0.3384,
     "answered": 198,
     "items": 200
    },
    "OpenJev": {
     "decisionScore": 74.69,
     "accuracy": 0.665,
     "answered": 200,
     "items": 200
    },
    "System One Scorer": {
     "decisionScore": 69.53,
     "accuracy": 0.515,
     "answered": 200,
     "items": 200
    }
   }
  },
  {
   "task": "cfpb-complaint-product-queue",
   "family": "workflow-control",
   "use_case": "support-inbox-triage",
   "vertical": "finance",
   "dataset": "liri-uzh/cfpb-complaints-mini",
   "licence": "CC0-1.0 (CFPB public domain)",
   "provenance": "outcome",
   "primitive": "choice",
   "items": 199,
   "scores": {
    "decider-2b": {
     "decisionScore": 79.71,
     "accuracy": 0.7286,
     "answered": 199,
     "items": 199
    },
    "Jev": {
     "decisionScore": 83.45,
     "accuracy": 0.794,
     "answered": 199,
     "items": 199
    },
    "JevFish": {
     "decisionScore": 78.76,
     "accuracy": 0.7136,
     "answered": 199,
     "items": 199
    },
    "NanoJev": {
     "decisionScore": 63.41,
     "accuracy": 0.4372,
     "answered": 199,
     "items": 199
    },
    "OpenJev": {
     "decisionScore": 78.94,
     "accuracy": 0.7688,
     "answered": 199,
     "items": 199
    },
    "System One Scorer": {
     "decisionScore": 82.62,
     "accuracy": 0.7789,
     "answered": 199,
     "items": 199
    }
   }
  },
  {
   "task": "cuad-checklist",
   "family": "safety-and-quality",
   "use_case": "compliance-verification",
   "vertical": "legal",
   "dataset": "nguha/legalbench",
   "licence": "CC-BY-4.0",
   "provenance": "expert",
   "primitive": "noul",
   "items": 204,
   "scores": {
    "decider-2b": {
     "decisionScore": 95.1,
     "accuracy": 0.951,
     "answered": 204,
     "items": 204
    },
    "Jev": {
     "decisionScore": 91.61,
     "accuracy": 0.8922,
     "answered": 204,
     "items": 204
    },
    "JevFish": {
     "decisionScore": 93.6,
     "accuracy": 0.9314,
     "answered": 204,
     "items": 204
    },
    "NanoJev": {
     "decisionScore": 72.65,
     "accuracy": 0.4951,
     "answered": 204,
     "items": 204
    },
    "OpenJev": {
     "decisionScore": 91.22,
     "accuracy": 0.8922,
     "answered": 204,
     "items": 204
    },
    "System One Scorer": {
     "decisionScore": 92.1,
     "accuracy": 0.9265,
     "answered": 204,
     "items": 204
    }
   }
  },
  {
   "task": "dbpedia-ontology-hierarchy",
   "family": "data-and-operations",
   "use_case": "hierarchical-classification",
   "vertical": "media-and-reference",
   "dataset": "DeveloperOats/DBPedia_Classes",
   "licence": "CC0-1.0",
   "provenance": "outcome",
   "primitive": "choice",
   "items": 200,
   "scores": {
    "decider-2b": {
     "decisionScore": 89.04,
     "accuracy": 0.89,
     "answered": 200,
     "items": 200
    },
    "Jev": {
     "decisionScore": 92.86,
     "accuracy": 0.94,
     "answered": 200,
     "items": 200
    },
    "JevFish": {
     "decisionScore": 89.58,
     "accuracy": 0.9,
     "answered": 200,
     "items": 200
    },
    "NanoJev": {
     "decisionScore": 60.64,
     "accuracy": 0.155,
     "answered": 200,
     "items": 200
    },
    "OpenJev": {
     "decisionScore": 89.31,
     "accuracy": 0.925,
     "answered": 200,
     "items": 200
    },
    "System One Scorer": {
     "decisionScore": 83.82,
     "accuracy": 0.87,
     "answered": 200,
     "items": 200
    }
   }
  },
  {
   "task": "esci-product-rerank",
   "family": "retrieval-and-knowledge",
   "use_case": "semantic-reranking",
   "vertical": "e-commerce",
   "dataset": "tasksource/esci",
   "licence": "Apache-2.0",
   "provenance": "crowd",
   "primitive": "noul",
   "items": 490,
   "scores": {
    "decider-2b": {
     "decisionScore": 74.15,
     "accuracy": 0.6204,
     "answered": 490,
     "items": 490
    },
    "Jev": {
     "decisionScore": 74.99,
     "accuracy": 0.7306,
     "answered": 490,
     "items": 490
    },
    "JevFish": {
     "decisionScore": 74.41,
     "accuracy": 0.6306,
     "answered": 490,
     "items": 490
    },
    "NanoJev": {
     "decisionScore": 71.08,
     "accuracy": 0.593,
     "answered": 489,
     "items": 490
    },
    "OpenJev": {
     "decisionScore": 71.82,
     "accuracy": 0.7224,
     "answered": 490,
     "items": 490
    },
    "System One Scorer": {
     "decisionScore": 75.51,
     "accuracy": 0.6163,
     "answered": 490,
     "items": 490
    }
   }
  },
  {
   "task": "finer-xbrl-fact-tag",
   "family": "data-and-operations",
   "use_case": "structured-extraction",
   "vertical": "finance",
   "dataset": "nlpaueb/finer-139",
   "licence": "CC-BY-SA-4.0",
   "provenance": "expert",
   "primitive": "choice",
   "items": 181,
   "scores": {
    "decider-2b": {
     "decisionScore": 79.81,
     "accuracy": 0.7293,
     "answered": 181,
     "items": 181
    },
    "Jev": {
     "decisionScore": 93.3,
     "accuracy": 0.9227,
     "answered": 181,
     "items": 181
    },
    "JevFish": {
     "decisionScore": 82.71,
     "accuracy": 0.7735,
     "answered": 181,
     "items": 181
    },
    "NanoJev": {
     "decisionScore": 53.03,
     "accuracy": 0.1381,
     "answered": 181,
     "items": 181
    },
    "OpenJev": {
     "decisionScore": 88.96,
     "accuracy": 0.8508,
     "answered": 181,
     "items": 181
    },
    "System One Scorer": {
     "decisionScore": 79.57,
     "accuracy": 0.8011,
     "answered": 181,
     "items": 181
    }
   }
  },
  {
   "task": "hotpot-context-filter",
   "family": "retrieval-and-knowledge",
   "use_case": "rag-passage-filtering",
   "vertical": "knowledge-work",
   "dataset": "hotpotqa/hotpot_qa",
   "licence": "CC-BY-SA-4.0",
   "provenance": "crowd",
   "primitive": "noul",
   "items": 200,
   "scores": {
    "decider-2b": {
     "decisionScore": 86.67,
     "accuracy": 0.84,
     "answered": 200,
     "items": 200
    },
    "Jev": {
     "decisionScore": 96.3,
     "accuracy": 0.945,
     "answered": 200,
     "items": 200
    },
    "JevFish": {
     "decisionScore": 90.67,
     "accuracy": 0.825,
     "answered": 200,
     "items": 200
    },
    "NanoJev": {
     "decisionScore": 0.0,
     "accuracy": null,
     "answered": 0,
     "items": 200
    },
    "OpenJev": {
     "decisionScore": 85.71,
     "accuracy": 0.93,
     "answered": 200,
     "items": 200
    },
    "System One Scorer": {
     "decisionScore": 75.4,
     "accuracy": 0.59,
     "answered": 200,
     "items": 200
    }
   }
  },
  {
   "task": "humanevalpack-spec-match",
   "family": "safety-and-quality",
   "use_case": "semantic-linting",
   "vertical": "software-engineering",
   "dataset": "bigcode/humanevalpack",
   "licence": "MIT",
   "provenance": "verifier",
   "primitive": "noul",
   "items": 164,
   "scores": {
    "decider-2b": {
     "decisionScore": 71.39,
     "accuracy": 0.6524,
     "answered": 164,
     "items": 164
    },
    "Jev": {
     "decisionScore": 82.84,
     "accuracy": 0.8902,
     "answered": 164,
     "items": 164
    },
    "JevFish": {
     "decisionScore": 71.62,
     "accuracy": 0.628,
     "answered": 164,
     "items": 164
    },
    "NanoJev": {
     "decisionScore": 68.07,
     "accuracy": 0.5031,
     "answered": 163,
     "items": 164
    },
    "OpenJev": {
     "decisionScore": 80.53,
     "accuracy": 0.8841,
     "answered": 164,
     "items": 164
    },
    "System One Scorer": {
     "decisionScore": 73.16,
     "accuracy": 0.6585,
     "answered": 164,
     "items": 164
    }
   }
  },
  {
   "task": "hwu64-tool-select",
   "family": "workflow-control",
   "use_case": "typed-tool-dispatch",
   "vertical": "consumer-assistant",
   "dataset": "deutsche-telekom/NLU-Evaluation-Data-en-de",
   "licence": "CC-BY-4.0",
   "provenance": "crowd",
   "primitive": "choice",
   "items": 200,
   "scores": {
    "decider-2b": {
     "decisionScore": 91.17,
     "accuracy": 0.86,
     "answered": 200,
     "items": 200
    },
    "Jev": {
     "decisionScore": 93.66,
     "accuracy": 0.88,
     "answered": 200,
     "items": 200
    },
    "JevFish": {
     "decisionScore": 91.85,
     "accuracy": 0.88,
     "answered": 200,
     "items": 200
    },
    "NanoJev": {
     "decisionScore": 67.8,
     "accuracy": 0.36,
     "answered": 200,
     "items": 200
    },
    "OpenJev": {
     "decisionScore": 91.45,
     "accuracy": 0.87,
     "answered": 200,
     "items": 200
    },
    "System One Scorer": {
     "decisionScore": 83.37,
     "accuracy": 0.805,
     "answered": 200,
     "items": 200
    }
   }
  },
  {
   "task": "k8s-issue-kind-routing",
   "family": "real-time-and-agents",
   "use_case": "agent-routing-and-skill-selection",
   "vertical": "devops",
   "dataset": "open-index/open-github-issues",
   "licence": "ODC-BY-1.0",
   "provenance": "outcome",
   "primitive": "choice",
   "items": 200,
   "scores": {
    "decider-2b": {
     "decisionScore": 75.36,
     "accuracy": 0.66,
     "answered": 200,
     "items": 200
    },
    "Jev": {
     "decisionScore": 77.46,
     "accuracy": 0.725,
     "answered": 200,
     "items": 200
    },
    "JevFish": {
     "decisionScore": 75.54,
     "accuracy": 0.645,
     "answered": 200,
     "items": 200
    },
    "NanoJev": {
     "decisionScore": 61.47,
     "accuracy": 0.34,
     "answered": 200,
     "items": 200
    },
    "OpenJev": {
     "decisionScore": 73.84,
     "accuracy": 0.71,
     "answered": 200,
     "items": 200
    },
    "System One Scorer": {
     "decisionScore": 73.46,
     "accuracy": 0.61,
     "answered": 200,
     "items": 200
    }
   }
  },
  {
   "task": "ledgar-clause-review-routing",
   "family": "workflow-control",
   "use_case": "intent-and-model-routing",
   "vertical": "legal",
   "dataset": "coastalcph/lex_glue",
   "licence": "CC-BY-4.0",
   "provenance": "outcome",
   "primitive": "choice",
   "items": 199,
   "scores": {
    "decider-2b": {
     "decisionScore": 37.3,
     "accuracy": 0.2513,
     "answered": 199,
     "items": 199
    },
    "Jev": {
     "decisionScore": 34.55,
     "accuracy": 0.2864,
     "answered": 199,
     "items": 199
    },
    "JevFish": {
     "decisionScore": 41.45,
     "accuracy": 0.2663,
     "answered": 199,
     "items": 199
    },
    "NanoJev": {
     "decisionScore": 50.8,
     "accuracy": 0.1407,
     "answered": 199,
     "items": 199
    },
    "OpenJev": {
     "decisionScore": 32.75,
     "accuracy": 0.2814,
     "answered": 199,
     "items": 199
    },
    "System One Scorer": {
     "decisionScore": 48.89,
     "accuracy": 0.201,
     "answered": 199,
     "items": 199
    }
   }
  },
  {
   "task": "moderation-hate-severity",
   "family": "safety-and-quality",
   "use_case": "llm-guardrails",
   "vertical": "trust-and-safety",
   "dataset": "mmathys/openai-moderation-api-evaluation",
   "licence": "MIT",
   "provenance": "expert",
   "primitive": "score",
   "items": 200,
   "scores": {
    "decider-2b": {
     "decisionScore": 89.64,
     "accuracy": 0.78,
     "answered": 200,
     "items": 200
    },
    "Jev": {
     "decisionScore": 93.38,
     "accuracy": 0.85,
     "answered": 200,
     "items": 200
    },
    "JevFish": {
     "decisionScore": 90.08,
     "accuracy": 0.79,
     "answered": 200,
     "items": 200
    },
    "NanoJev": {
     "decisionScore": 75.03,
     "accuracy": 0.35,
     "answered": 200,
     "items": 200
    },
    "OpenJev": {
     "decisionScore": 90.31,
     "accuracy": 0.775,
     "answered": 200,
     "items": 200
    },
    "System One Scorer": {
     "decisionScore": 86.32,
     "accuracy": 0.74,
     "answered": 200,
     "items": 200
    }
   }
  },
  {
   "task": "prompt-injection",
   "family": "safety-and-quality",
   "use_case": "llm-guardrails",
   "vertical": "security",
   "dataset": "deepset/prompt-injections",
   "licence": "Apache-2.0",
   "provenance": "crowd",
   "primitive": "noul",
   "items": 194,
   "scores": {
    "decider-2b": {
     "decisionScore": 76.3,
     "accuracy": 0.701,
     "answered": 194,
     "items": 194
    },
    "Jev": {
     "decisionScore": 90.54,
     "accuracy": 0.8763,
     "answered": 194,
     "items": 194
    },
    "JevFish": {
     "decisionScore": 79.12,
     "accuracy": 0.7165,
     "answered": 194,
     "items": 194
    },
    "NanoJev": {
     "decisionScore": 77.92,
     "accuracy": 0.6443,
     "answered": 194,
     "items": 194
    },
    "OpenJev": {
     "decisionScore": 85.38,
     "accuracy": 0.8351,
     "answered": 194,
     "items": 194
    },
    "System One Scorer": {
     "decisionScore": 82.43,
     "accuracy": 0.7577,
     "answered": 194,
     "items": 194
    }
   }
  },
  {
   "task": "swe-rebench-run-resolved",
   "family": "real-time-and-agents",
   "use_case": "run-judging",
   "vertical": "software-engineering",
   "dataset": "nebius/SWE-rebench-openhands-trajectories",
   "licence": "CC-BY-4.0",
   "provenance": "verifier",
   "primitive": "noul",
   "items": 200,
   "scores": {
    "decider-2b": {
     "decisionScore": 69.39,
     "accuracy": 0.52,
     "answered": 200,
     "items": 200
    },
    "Jev": {
     "decisionScore": 74.88,
     "accuracy": 0.6,
     "answered": 200,
     "items": 200
    },
    "JevFish": {
     "decisionScore": 75.3,
     "accuracy": 0.565,
     "answered": 200,
     "items": 200
    },
    "NanoJev": {
     "decisionScore": 0.0,
     "accuracy": null,
     "answered": 0,
     "items": 200
    },
    "OpenJev": {
     "decisionScore": 59.23,
     "accuracy": 0.555,
     "answered": 200,
     "items": 200
    },
    "System One Scorer": {
     "decisionScore": 73.45,
     "accuracy": 0.5,
     "answered": 200,
     "items": 200
    }
   }
  },
  {
   "task": "swe-smith-run-resolved",
   "family": "real-time-and-agents",
   "use_case": "run-judging",
   "vertical": "software-engineering",
   "dataset": "SWE-bench/SWE-smith-trajectories",
   "licence": "MIT",
   "provenance": "verifier",
   "primitive": "noul",
   "items": 200,
   "scores": {
    "decider-2b": {
     "decisionScore": 69.76,
     "accuracy": 0.46,
     "answered": 200,
     "items": 200
    },
    "Jev": {
     "decisionScore": 60.2,
     "accuracy": 0.505,
     "answered": 200,
     "items": 200
    },
    "JevFish": {
     "decisionScore": 63.74,
     "accuracy": 0.535,
     "answered": 200,
     "items": 200
    },
    "NanoJev": {
     "decisionScore": 0.33,
     "accuracy": 0.0,
     "answered": 1,
     "items": 200
    },
    "OpenJev": {
     "decisionScore": 51.14,
     "accuracy": 0.495,
     "answered": 200,
     "items": 200
    },
    "System One Scorer": {
     "decisionScore": 71.84,
     "accuracy": 0.5,
     "answered": 200,
     "items": 200
    }
   }
  },
  {
   "task": "tau2-airline-run-success",
   "family": "real-time-and-agents",
   "use_case": "run-judging",
   "vertical": "customer-operations",
   "dataset": "snorkelai/Tau2-Bench-Verified-Airline-With-Code-Agents",
   "licence": "Apache-2.0",
   "provenance": "verifier",
   "primitive": "noul",
   "items": 200,
   "scores": {
    "decider-2b": {
     "decisionScore": 67.32,
     "accuracy": 0.47,
     "answered": 200,
     "items": 200
    },
    "Jev": {
     "decisionScore": 68.02,
     "accuracy": 0.52,
     "answered": 200,
     "items": 200
    },
    "JevFish": {
     "decisionScore": 64.89,
     "accuracy": 0.515,
     "answered": 200,
     "items": 200
    },
    "NanoJev": {
     "decisionScore": 75.47,
     "accuracy": 0.5,
     "answered": 198,
     "items": 200
    },
    "OpenJev": {
     "decisionScore": 55.26,
     "accuracy": 0.475,
     "answered": 200,
     "items": 200
    },
    "System One Scorer": {
     "decisionScore": 71.85,
     "accuracy": 0.485,
     "answered": 200,
     "items": 200
    }
   }
  },
  {
   "task": "tau2-airline-transcript-success",
   "family": "real-time-and-agents",
   "use_case": "run-judging",
   "vertical": "customer-operations",
   "dataset": "snorkelai/Tau2-Bench-Verified-Airline-With-Code-Agents",
   "licence": "Apache-2.0",
   "provenance": "verifier",
   "primitive": "noul",
   "items": 199,
   "scores": {
    "decider-2b": {
     "decisionScore": 63.88,
     "accuracy": 0.3166,
     "answered": 199,
     "items": 199
    },
    "Jev": {
     "decisionScore": 76.74,
     "accuracy": 0.608,
     "answered": 199,
     "items": 199
    },
    "JevFish": {
     "decisionScore": 65.67,
     "accuracy": 0.3668,
     "answered": 199,
     "items": 199
    },
    "NanoJev": {
     "decisionScore": 1.36,
     "accuracy": 0.6667,
     "answered": 3,
     "items": 199
    },
    "OpenJev": {
     "decisionScore": 63.68,
     "accuracy": 0.4874,
     "answered": 199,
     "items": 199
    },
    "System One Scorer": {
     "decisionScore": 74.83,
     "accuracy": 0.5276,
     "answered": 199,
     "items": 199
    }
   }
  },
  {
   "task": "trial-outcomes",
   "family": "data-and-operations",
   "use_case": "feature-extraction-for-ml",
   "vertical": "saas-growth",
   "dataset": "synthetic:saas-trial-v1",
   "licence": "CC-BY-4.0 (generated here)",
   "provenance": "oracle",
   "primitive": "noul",
   "items": 400,
   "scores": {
    "decider-2b": {
     "decisionScore": 77.85,
     "accuracy": 0.695,
     "answered": 400,
     "items": 400
    },
    "Jev": {
     "decisionScore": 78.75,
     "accuracy": 0.6775,
     "answered": 400,
     "items": 400
    },
    "JevFish": {
     "decisionScore": 75.35,
     "accuracy": 0.695,
     "answered": 400,
     "items": 400
    },
    "NanoJev": {
     "decisionScore": 76.4,
     "accuracy": 0.695,
     "answered": 400,
     "items": 400
    },
    "OpenJev": {
     "decisionScore": 69.41,
     "accuracy": 0.6925,
     "answered": 400,
     "items": 400
    },
    "System One Scorer": {
     "decisionScore": 77.74,
     "accuracy": 0.695,
     "answered": 400,
     "items": 400
    }
   }
  },
  {
   "task": "trialgpt-criterion-support",
   "family": "retrieval-and-knowledge",
   "use_case": "citation-verification",
   "vertical": "healthcare",
   "dataset": "ncbi/TrialGPT-Criterion-Annotations",
   "licence": "Public domain (NCBI)",
   "provenance": "expert",
   "primitive": "choice",
   "items": 159,
   "scores": {
    "decider-2b": {
     "decisionScore": 73.31,
     "accuracy": 0.5786,
     "answered": 159,
     "items": 159
    },
    "Jev": {
     "decisionScore": 73.79,
     "accuracy": 0.6415,
     "answered": 159,
     "items": 159
    },
    "JevFish": {
     "decisionScore": 73.51,
     "accuracy": 0.6164,
     "answered": 159,
     "items": 159
    },
    "NanoJev": {
     "decisionScore": 58.71,
     "accuracy": 0.1258,
     "answered": 159,
     "items": 159
    },
    "OpenJev": {
     "decisionScore": 60.57,
     "accuracy": 0.5346,
     "answered": 159,
     "items": 159
    },
    "System One Scorer": {
     "decisionScore": 70.57,
     "accuracy": 0.5346,
     "answered": 159,
     "items": 159
    }
   }
  },
  {
   "task": "unfair-tos-clause-type",
   "family": "safety-and-quality",
   "use_case": "compliance-verification",
   "vertical": "consumer-legal",
   "dataset": "nguha/legalbench",
   "licence": "CC-BY-4.0",
   "provenance": "expert",
   "primitive": "choice",
   "items": 200,
   "scores": {
    "decider-2b": {
     "decisionScore": 83.53,
     "accuracy": 0.8,
     "answered": 200,
     "items": 200
    },
    "Jev": {
     "decisionScore": 85.91,
     "accuracy": 0.82,
     "answered": 200,
     "items": 200
    },
    "JevFish": {
     "decisionScore": 87.26,
     "accuracy": 0.835,
     "answered": 200,
     "items": 200
    },
    "NanoJev": {
     "decisionScore": 60.83,
     "accuracy": 0.255,
     "answered": 200,
     "items": 200
    },
    "OpenJev": {
     "decisionScore": 83.92,
     "accuracy": 0.805,
     "answered": 200,
     "items": 200
    },
    "System One Scorer": {
     "decisionScore": 79.07,
     "accuracy": 0.78,
     "answered": 200,
     "items": 200
    }
   }
  },
  {
   "task": "unfair-tos-publish-gate",
   "family": "workflow-control",
   "use_case": "confidence-gated-actions",
   "vertical": "consumer-legal",
   "dataset": "coastalcph/lex_glue",
   "licence": "CC-BY-4.0",
   "provenance": "expert",
   "primitive": "noul",
   "items": 198,
   "scores": {
    "decider-2b": {
     "decisionScore": 83.27,
     "accuracy": 0.7576,
     "answered": 198,
     "items": 198
    },
    "Jev": {
     "decisionScore": 87.1,
     "accuracy": 0.8434,
     "answered": 198,
     "items": 198
    },
    "JevFish": {
     "decisionScore": 83.13,
     "accuracy": 0.7071,
     "answered": 198,
     "items": 198
    },
    "NanoJev": {
     "decisionScore": 75.97,
     "accuracy": 0.6263,
     "answered": 198,
     "items": 198
    },
    "OpenJev": {
     "decisionScore": 83.9,
     "accuracy": 0.8182,
     "answered": 198,
     "items": 198
    },
    "System One Scorer": {
     "decisionScore": 83.92,
     "accuracy": 0.8232,
     "answered": 198,
     "items": 198
    }
   }
  },
  {
   "task": "vitaminc-claim-consistency",
   "family": "safety-and-quality",
   "use_case": "self-consistency-checks",
   "vertical": "media",
   "dataset": "tals/vitaminc",
   "licence": "CC-BY-SA-3.0",
   "provenance": "crowd",
   "primitive": "choice",
   "items": 200,
   "scores": {
    "decider-2b": {
     "decisionScore": 82.42,
     "accuracy": 0.715,
     "answered": 200,
     "items": 200
    },
    "Jev": {
     "decisionScore": 88.08,
     "accuracy": 0.83,
     "answered": 200,
     "items": 200
    },
    "JevFish": {
     "decisionScore": 83.26,
     "accuracy": 0.73,
     "answered": 200,
     "items": 200
    },
    "NanoJev": {
     "decisionScore": 60.8,
     "accuracy": 0.505,
     "answered": 200,
     "items": 200
    },
    "OpenJev": {
     "decisionScore": 82.77,
     "accuracy": 0.77,
     "answered": 200,
     "items": 200
    },
    "System One Scorer": {
     "decisionScore": 82.32,
     "accuracy": 0.76,
     "answered": 200,
     "items": 200
    }
   }
  }
 ]
}
