{
  "schema_version": 2,
  "method": "evopilot",
  "dataset": "MovieLens 100K",
  "seeds": [
    42,
    43,
    44
  ],
  "runs": [
    {
      "seed": 42,
      "baseline_hit_at_10": 0.10909090909090909,
      "baseline_ndcg_at_10": 0.05400546778896325,
      "baseline_fresh_hit_at_10": 0.031746031746031744,
      "baseline_head_share_at_10": 0.23772727272727273,
      "method_hit_at_10": 0.10454545454545454,
      "method_ndcg_at_10": 0.05078228660346273,
      "method_fresh_hit_at_10": 0.031746031746031744,
      "method_head_share_at_10": 0.19590909090909092,
      "hit_at_10_relative_percent": -4.1666666666666625,
      "validation_only_alpha": 0.1,
      "historical_mismatch_rejected": true,
      "matched_comparison_admitted": true
    },
    {
      "seed": 43,
      "baseline_hit_at_10": 0.10909090909090909,
      "baseline_ndcg_at_10": 0.05400546778896325,
      "baseline_fresh_hit_at_10": 0.031746031746031744,
      "baseline_head_share_at_10": 0.23772727272727273,
      "method_hit_at_10": 0.10454545454545454,
      "method_ndcg_at_10": 0.05078228660346273,
      "method_fresh_hit_at_10": 0.031746031746031744,
      "method_head_share_at_10": 0.19590909090909092,
      "hit_at_10_relative_percent": -4.1666666666666625,
      "validation_only_alpha": 0.1,
      "historical_mismatch_rejected": true,
      "matched_comparison_admitted": true
    },
    {
      "seed": 44,
      "baseline_hit_at_10": 0.10909090909090909,
      "baseline_ndcg_at_10": 0.05400546778896325,
      "baseline_fresh_hit_at_10": 0.031746031746031744,
      "baseline_head_share_at_10": 0.23772727272727273,
      "method_hit_at_10": 0.10454545454545454,
      "method_ndcg_at_10": 0.05078228660346273,
      "method_fresh_hit_at_10": 0.031746031746031744,
      "method_head_share_at_10": 0.19590909090909092,
      "hit_at_10_relative_percent": -4.1666666666666625,
      "validation_only_alpha": 0.1,
      "historical_mismatch_rejected": true,
      "matched_comparison_admitted": true
    }
  ],
  "aggregate_metrics": {
    "baseline_fresh_hit_at_10_mean": 0.031746031746031744,
    "baseline_fresh_hit_at_10_std": 0.0,
    "baseline_head_share_at_10_mean": 0.2377272727272727,
    "baseline_head_share_at_10_std": 0.0,
    "baseline_hit_at_10_mean": 0.10909090909090909,
    "baseline_hit_at_10_std": 0.0,
    "baseline_ndcg_at_10_mean": 0.05400546778896325,
    "baseline_ndcg_at_10_std": 0.0,
    "hit_at_10_relative_percent_mean": -4.1666666666666625,
    "hit_at_10_relative_percent_std": 0.0,
    "method_fresh_hit_at_10_mean": 0.031746031746031744,
    "method_fresh_hit_at_10_std": 0.0,
    "method_head_share_at_10_mean": 0.19590909090909092,
    "method_head_share_at_10_std": 0.0,
    "method_hit_at_10_mean": 0.10454545454545454,
    "method_hit_at_10_std": 0.0,
    "method_ndcg_at_10_mean": 0.05078228660346273,
    "method_ndcg_at_10_std": 0.0,
    "seed_mean": 43.0,
    "seed_std": 1.0,
    "validation_only_alpha_mean": 0.10000000000000002,
    "validation_only_alpha_std": 0.0
  },
  "manifest_ref": "recommendation:evopilot",
  "evaluation_protocol": {
    "tier": "l2_public_dataset",
    "seeds": [
      42,
      43,
      44
    ],
    "formal_comparison": true,
    "diagnostic_only": false,
    "claim_policy": "matched public comparison; negative results retained; paper-scale and production VDD claims excluded"
  },
  "provenance": {
    "commit": "working tree before commit",
    "command": "PYTHONPATH=src python scripts/generate_sep_21_agent_artifacts.py",
    "artifact_path": "docs/reproductions/2609.21257-evopilot/metrics/movielens-100k-seeds42-44.json",
    "dataset_fingerprint": "movielens-100k-compact-temporal-v1"
  }
}
