{
  "schema_version": 2,
  "manifest_ref": "agent-research:convapparel",
  "adapter_key": "convapparel",
  "evaluation_tier": "l0_interface",
  "diagnostic_only": true,
  "promotion_eligible": false,
  "evolve_integrated": false,
  "formal_comparison": false,
  "seed": 42,
  "device": "cpu",
  "dataset": {
    "id": "self-authored-protocol-fixtures",
    "public_benchmark": false,
    "revision": "google16-evaluation-fixtures-v1"
  },
  "metrics": {
    "conversations": 8,
    "simulated_hls": 0.14582349109712062,
    "human_hls": 0.8541765089028793,
    "discriminator_accuracy": 1,
    "brier_score": 0.021264490555752025
  },
  "full_experiment_status": "not_reproduced_high_cost",
  "evaluation_protocol": {
    "tier": "l0_interface",
    "seeds": [
      42
    ],
    "formal_comparison": false,
    "claim_policy": "bounded component/protocol only, not paper capability"
  },
  "provenance": {
    "artifact_path": "docs/agent-research/2602.16938-convapparel/metrics/protocol-cpu-seed42.json",
    "code_commit": "6873b033ffa97bee066d08436b16fac53bb42ccc",
    "dataset_fingerprint": "bef370cfe12f2cebda1bc612b7b20cd679a4227b8d2e48e7cd7f3b050f2ba7fb",
    "raw_result_path": "docs/research-audits/google16-components-20261021/evidence/google16-convapparel-final-seed42.json",
    "raw_result_sha256": "050f1cd36e29b33ea2eaa024bfb55047898ac353ef0595d14e37f400b2a064dc",
    "source_binding": "curator_verified_frozen_files"
  },
  "fixture_demo": true,
  "model_calls": 0,
  "report": {
    "training_policy": "good",
    "target_policy": "bad",
    "simulator_training_provenance": "declared_manifest_not_independently_certified",
    "plsa": {
      "satisfaction": {
        "human_mean": 0.6,
        "simulated_mean": 0.8,
        "ks": 1.0,
        "ks_pvalue": 0.02857142857142857,
        "mann_whitney_u": 0.0,
        "mwu_pvalue": 0.02857142857142857,
        "pvalue_method": "exact_permutation",
        "permutations": 70
      },
      "turns_per_session": {
        "human_mean": 1.0,
        "simulated_mean": 1.0,
        "ks": 0.0,
        "ks_pvalue": 1.0,
        "mann_whitney_u": 8.0,
        "mwu_pvalue": 1.0,
        "pvalue_method": "exact_permutation",
        "permutations": 70
      },
      "words_per_user_turn": {
        "human_mean": 4.0,
        "simulated_mean": 4.0,
        "ks": 0.0,
        "ks_pvalue": 1.0,
        "mann_whitney_u": 8.0,
        "mwu_pvalue": 1.0,
        "pvalue_method": "exact_permutation",
        "permutations": 70
      }
    },
    "hls": {
      "conversations": 8,
      "simulated_hls": 0.14582349109712062,
      "human_hls": 0.8541765089028793,
      "discriminator_accuracy": 1,
      "brier_score": 0.021264490555752025
    },
    "diagnostic_only": true,
    "discriminator_backend": "cpu_bag_of_words_logistic_not_paper_gemini"
  },
  "status": "completed"
}