{ "bench_version": 1, "track": "tabular", "model": "mlp_tiny", "metric_direction": "lower is better", "n_seeds": 8, "baseline": { "best_cfg": { "lr": 0.024, "tau": 1.0 }, "sweep": [ { "cfg": { "lr": 0.0015, "tau": 0.5 }, "mean": 149.69311714172363 }, { "cfg": { "lr": 0.0015, "tau": 1.0 }, "mean": 149.69156074523926 }, { "cfg": { "lr": 0.0015, "tau": 2.0 }, "mean": 149.69087982177734 }, { "cfg": { "lr": 0.003, "tau": 0.5 }, "mean": 20.230180501937866 }, { "cfg": { "lr": 0.003, "tau": 1.0 }, "mean": 20.230008840560913 }, { "cfg": { "lr": 0.003, "tau": 2.0 }, "mean": 20.22838020324707 }, { "cfg": { "lr": 0.006, "tau": 0.5 }, "mean": 17.811020731925964 }, { "cfg": { "lr": 0.006, "tau": 1.0 }, "mean": 17.814178228378296 }, { "cfg": { "lr": 0.006, "tau": 2.0 }, "mean": 17.815396189689636 }, { "cfg": { "lr": 0.012, "tau": 0.5 }, "mean": 13.617882013320923 }, { "cfg": { "lr": 0.012, "tau": 1.0 }, "mean": 13.623129487037659 }, { "cfg": { "lr": 0.012, "tau": 2.0 }, "mean": 13.62790322303772 }, { "cfg": { "lr": 0.024, "tau": 0.5 }, "mean": 11.456546604633331 }, { "cfg": { "lr": 0.024, "tau": 1.0 }, "mean": 11.436318755149841 }, { "cfg": { "lr": 0.024, "tau": 2.0 }, "mean": 11.445708155632019 } ], "full": { "mean": 11.436318755149841, "std": 2.802007267234235, "per_seed": [ 13.539134979248047, 14.802740097045898, 14.924714088439941, 9.829378128051758, 7.340484619140625, 7.899506568908691, 10.356738090515137, 12.797853469848633 ], "n": 8 } }, "idea": { "mean": 15.096654891967773, "std": 1.5912454314949351, "per_seed": [ 14.8751220703125, 15.759033203125, 13.488835334777832, 14.056472778320312, 13.784052848815918, 13.798114776611328, 16.769433975219727, 18.24217414855957 ], "n": 8, "cfg": { "lr": 0.024, "tau": 1.0 } }, "comparison": { "delta_mean": 3.660336136817932, "idea_wins": 1, "n_pairs": 8, "per_seed_diffs": [ 1.3359870910644531, 0.9562931060791016, -1.4358787536621094, 4.227094650268555, 6.443568229675293, 5.898608207702637, 6.41269588470459, 5.4443206787109375 ], "p_value": 0.03245, "mde": 2.498914897900942, "mde_rel_pct": 21.85069296687508, "verdict": "idea worse (significant)", "system_worked": false }, "idea_grid": [ { "mean": 15.096654891967773, "std": 1.5912454314949351, "per_seed": [ 14.8751220703125, 15.759033203125, 13.488835334777832, 14.056472778320312, 13.784052848815918, 13.798114776611328, 16.769433975219727, 18.24217414855957 ], "n": 8, "cfg": { "lr": 0.024, "tau": 1.0 } }, { "mean": 22.211764812469482, "std": 2.486166287168737, "per_seed": [ 22.993654251098633, 21.551036834716797, 27.089385986328125, 18.64665412902832, 22.97047233581543, 18.93794059753418, 22.85433006286621, 22.650644302368164 ], "n": 8, "cfg": { "lr": 0.006, "tau": 1.0 } }, { "mean": 15.096654891967773, "std": 1.5912454314949351, "per_seed": [ 14.8751220703125, 15.759033203125, 13.488835334777832, 14.056472778320312, 13.784052848815918, 13.798114776611328, 16.769433975219727, 18.24217414855957 ], "n": 8, "cfg": { "lr": 0.024, "tau": 1.0 } } ], "paired_diffs_baseline_minus_idea": [ -1.3359870910644531, -0.9562931060791016, 1.4358787536621094, -4.227094650268555, -6.443568229675293, -5.898608207702637, -6.41269588470459, -5.4443206787109375 ], "permutation_pvalue": 0.03245, "mechanism_signature": { "prediction": "residual norm <= tau and common-direction cosine remains near 1", "predicted_residual_bound": 1.0, "observed_max_residual_mean": 1.0000003725290298, "observed_direction_cosine_mean": 0.5656541213393211, "confirmed": false, "per_seed": [ { "metric": 14.8751220703125, "observed_max_residual": 1.000000238418579, "predicted_tau": 1.0, "direction_cosine": 0.2306671440601349 }, { "metric": 15.759033203125, "observed_max_residual": 1.0000005960464478, "predicted_tau": 1.0, "direction_cosine": 0.8679510951042175 }, { "metric": 13.488835334777832, "observed_max_residual": 1.0000005960464478, "predicted_tau": 1.0, "direction_cosine": 0.8376753926277161 }, { "metric": 14.056472778320312, "observed_max_residual": 1.0000003576278687, "predicted_tau": 1.0, "direction_cosine": 0.16625230014324188 }, { "metric": 13.784052848815918, "observed_max_residual": 1.0000005960464478, "predicted_tau": 1.0, "direction_cosine": 0.8546982407569885 }, { "metric": 13.798114776611328, "observed_max_residual": 1.0000001192092896, "predicted_tau": 1.0, "direction_cosine": 0.8671970367431641 }, { "metric": 16.769433975219727, "observed_max_residual": 1.000000238418579, "predicted_tau": 1.0, "direction_cosine": 0.6522518992424011 }, { "metric": 18.24217414855957, "observed_max_residual": 1.000000238418579, "predicted_tau": 1.0, "direction_cosine": 0.04853986203670502 } ] } }