{
  "bench_version": 1,
  "track": "dynamics",
  "model": "rnn_small",
  "metric_direction": "lower is better",
  "n_seeds": 8,
  "baseline": {
    "best_cfg": {
      "lr": 0.006,
      "epochs": 12
    },
    "sweep": [
      {
        "cfg": {
          "lr": 0.001,
          "epochs": 12
        },
        "mean": 0.004298779065720737
      },
      {
        "cfg": {
          "lr": 0.003,
          "epochs": 12
        },
        "mean": 0.0034615940821822733
      },
      {
        "cfg": {
          "lr": 0.006,
          "epochs": 12
        },
        "mean": 0.0018282315868418664
      }
    ],
    "full": {
      "mean": 0.002105819425196387,
      "std": 0.000389436566226016,
      "per_seed": [
        0.002004395006224513,
        0.001437882543541491,
        0.0015748648438602686,
        0.002295783953741193,
        0.0024020641576498747,
        0.0024898042902350426,
        0.0020851334556937218,
        0.0025566271506249905
      ],
      "n": 8
    }
  },
  "idea": {
    "mean": 0.7633235901594162,
    "std": 0.05257852286312429,
    "per_seed": [
      0.746772825717926,
      0.7365002632141113,
      0.8016765713691711,
      0.886947512626648,
      0.735880434513092,
      0.7473666071891785,
      0.7420515418052673,
      0.7093929648399353
    ],
    "n": 8
  },
  "comparison": {
    "delta_mean": 0.7612177707342198,
    "idea_wins": 0,
    "n_pairs": 8,
    "per_seed_diffs": [
      0.7447684307117015,
      0.7350623806705698,
      0.8001017065253109,
      0.8846517286729068,
      0.7334783703554422,
      0.7448768028989434,
      0.7399664083495736,
      0.7068363376893103
    ],
    "p_value": 0.0081,
    "mde": 0.04703323792024009,
    "mde_rel_pct": 2233.4886532758533,
    "verdict": "idea worse (significant)",
    "system_worked": false
  },
  "mechanism_signature": {
    "prediction": "decision-aware training reduces surrogate optimum displacement relative to MSE-only while preserving trajectory fit",
    "measured_on_trained_models": true,
    "decision_probe": {
      "contexts": 32,
      "multistart": 5,
      "bounds": [
        -1.5,
        1.5
      ]
    },
    "confirmed": false,
    "note": "Standard task MSE is primary; probe is diagnostic only."
  },
  "idea_sweep": [
    {
      "cfg": {
        "lr": 0.001,
        "epochs": 12
      },
      "result": {
        "mean": 0.7633235901594162,
        "std": 0.05257852286312429,
        "per_seed": [
          0.746772825717926,
          0.7365002632141113,
          0.8016765713691711,
          0.886947512626648,
          0.735880434513092,
          0.7473666071891785,
          0.7420515418052673,
          0.7093929648399353
        ],
        "n": 8
      }
    },
    {
      "cfg": {
        "lr": 0.003,
        "epochs": 12
      },
      "result": {
        "mean": 0.7637511864304543,
        "std": 0.05106457675677259,
        "per_seed": [
          0.748903214931488,
          0.7398125529289246,
          0.802194356918335,
          0.8831914663314819,
          0.7348660230636597,
          0.7470683455467224,
          0.7429152727127075,
          0.7110582590103149
        ],
        "n": 8
      }
    },
    {
      "cfg": {
        "lr": 0.006,
        "epochs": 12
      },
      "result": {
        "mean": 0.7635712027549744,
        "std": 0.050801253101102256,
        "per_seed": [
          0.7507007122039795,
          0.7397510409355164,
          0.8021869659423828,
          0.8825896382331848,
          0.7341564297676086,
          0.7471616268157959,
          0.7382707595825195,
          0.7137524485588074
        ],
        "n": 8
      }
    }
  ],
  "selected_idea_cfg": {
    "lr": 0.001,
    "epochs": 12
  }
}