scripts/build_paper_table_status.py

#!/usr/bin/env python
from __future__ import annotations

import json
from dataclasses import asdict, dataclass
from pathlib import Path
from typing import Any


RESULTS_DIR = Path("results")
BASELINE_H16_POLICY = 0.29739130434782607


@dataclass(frozen=True)
class ResultSpec:
    key: str
    label: str
    path: str
    clean_deployment: str
    same_state_proposals: str
    expert_proposal: str
    story_role: str
    fallback_success: float | None = None
    pending_job: str = ""


SPECS = [
    ResultSpec(
        key="h16_policy",
        label="Direct h=16 policy",
        path="",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="behavior-cloning baseline",
        fallback_success=0.29739130434782607,
    ),
    ResultSpec(
        key="gaussian_field",
        label="Gaussian field search",
        path="h16_field_sweep_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="negative off-manifold field ablation",
        fallback_success=0.2910,
    ),
    ResultSpec(
        key="retrieval_lattice_no_expert",
        label="Nearest train-state lattice, no expert",
        path="h16_retrieval_lattice_no_expert_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="negative generic action-library ablation",
        fallback_success=0.2713,
    ),
    ResultSpec(
        key="near_miss_policy_bc5_field",
        label="Near-miss proposal policy + field",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_field_sweep_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="strong clean proposal-field bridge",
        fallback_success=0.3293,
    ),
    ResultSpec(
        key="field_optim",
        label="Trust-region field optimization",
        path="h16_field_optim_near_miss_policy_bc5_bestpt_s4_trust05_afterany_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="differentiable field-ascent diagnostic",
        pending_job="14842528/14842551",
    ),
    ResultSpec(
        key="nonexpert_policy_bc5",
        label="Best non-expert proposal policy",
        path="h16_policy_ckpt_nonexpert_policy_bc5_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="broader non-expert proposal-model ablation",
        pending_job="14842574/14842575/14842616",
    ),
    ResultSpec(
        key="nonexpert_policy_bc5_field",
        label="Best non-expert proposal policy + field",
        path="h16_policy_ckpt_nonexpert_policy_bc5_bestpt_field_sweep_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="broader proposal-field ablation",
        pending_job="14842574/14842577/14842617",
    ),
    ResultSpec(
        key="field_selected_noexpert_policy",
        label="Field-selected no-expert distillation policy",
        path="h16_policy_ckpt_field_selected_noexpert_bc5_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="student of field-on-lattice teacher",
        pending_job="14858327/14858328/14858329/14858330",
    ),
    ResultSpec(
        key="field_selected_noexpert_policy_field",
        label="Field-selected no-expert distillation + field",
        path="h16_policy_ckpt_field_selected_noexpert_bc5_bestpt_field_sweep_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="student proposal with field scoring",
        pending_job="14858327/14858328/14858331/14858332",
    ),
    ResultSpec(
        key="field_selected_noexpert_policy_allmap",
        label="Field-selected no-expert distillation policy, aligned validation",
        path="h16_policy_ckpt_field_selected_noexpert_bc5_allmap_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="field-teacher student with aligned checkpoint selection",
        pending_job="14858449/14858450/14858451/14858452",
    ),
    ResultSpec(
        key="field_selected_noexpert_policy_allmap_field",
        label="Field-selected no-expert distillation + field, aligned validation",
        path="h16_policy_ckpt_field_selected_noexpert_bc5_allmap_bestpt_field_sweep_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="aligned field-teacher student with field scoring",
        pending_job="14858449/14858450/14858453/14858454",
    ),
    ResultSpec(
        key="retrieval_residual_tangent_distill_allmap",
        label="Residual-tangent distillation policy, aligned validation",
        path="h16_policy_ckpt_residual_tangent_bc5_allmap_v2_best_policy_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="negative student of transported tangent teacher",
        pending_job="14862455/14862456/14862457/14862458",
    ),
    ResultSpec(
        key="retrieval_residual",
        label="Train-state counterfactual residual retrieval",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_v2_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="transferable local tangent proposal",
        pending_job="14857111/14857112/14857113",
    ),
    ResultSpec(
        key="retrieval_residual_scale025",
        label="Train-state residual retrieval, scale 0.25",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_scale0p25_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="tangent transport scale ablation",
        pending_job="14858875/14858876",
    ),
    ResultSpec(
        key="retrieval_residual_scale050",
        label="Train-state residual retrieval, scale 0.50",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_scale0p50_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="tangent transport scale ablation",
        pending_job="14858877/14858878",
    ),
    ResultSpec(
        key="retrieval_residual_scale050_zscore",
        label="Train-state residual retrieval, scale 0.50, z-score retrieval",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_scale0p50_zscore_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="state-normalized tangent retrieval ablation",
        pending_job="14859197/14859198",
    ),
    ResultSpec(
        key="retrieval_residual_scale050_zscore_no_random_wrongdir",
        label="Train-state residual retrieval, scale 0.50, z-score retrieval, no random/wrong-direction residuals",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_scale0p50_zscore_no_random_wrongdir_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="state-normalized typed tangent retrieval ablation",
        pending_job="14859199/14859200",
    ),
    ResultSpec(
        key="retrieval_residual_scale025_zscore_no_random_wrongdir",
        label="Train-state residual retrieval, scale 0.25, z-score retrieval, no random/wrong-direction residuals",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_scale0p25_zscore_no_random_wrongdir_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="state-normalized typed tangent retrieval ablation",
        pending_job="14859201/14859202",
    ),
    ResultSpec(
        key="retrieval_residual_scale050_no_random",
        label="Train-state residual retrieval, scale 0.50, no random residuals",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_scale0p50_no_random_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="anti-goal residual family mask ablation",
        pending_job="14859188/14859189",
    ),
    ResultSpec(
        key="retrieval_residual_scale050_no_random_wrongdir",
        label="Train-state residual retrieval, scale 0.50, no random/wrong-direction residuals",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_scale0p50_no_random_wrongdir_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="anti-goal residual family mask ablation",
        pending_job="14859191/14859192",
    ),
    ResultSpec(
        key="retrieval_residual_scale025_no_random_wrongdir",
        label="Train-state residual retrieval, scale 0.25, no random/wrong-direction residuals",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_scale0p25_no_random_wrongdir_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="anti-goal residual family mask ablation",
        pending_job="14859193/14859194",
    ),
    ResultSpec(
        key="retrieval_residual_scale050_safe_types",
        label="Train-state residual retrieval, scale 0.50, policy/no-op/wrong-gripper residuals",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_scale0p50_safe_types_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="typed tangent-family mask ablation",
        pending_job="14859195/14859196",
    ),
    ResultSpec(
        key="retrieval_residual_scale035_safe_types",
        label="Train-state residual retrieval, scale 0.35, policy/no-op/wrong-gripper residuals",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_scale0p35_safe_types_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="typed tangent scale fine sweep",
        pending_job="14859503/14859504",
    ),
    ResultSpec(
        key="retrieval_residual_scale035_safe_margin020",
        label="Train-state residual retrieval, scale 0.35, safe residuals, advantage margin 0.20",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_scale0p35_safe_types_margin0p20_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="counterfactual advantage abstention",
        pending_job="14862714/14862715",
    ),
    ResultSpec(
        key="retrieval_residual_scale050_safe_margin020",
        label="Train-state residual retrieval, scale 0.50, safe residuals, advantage margin 0.20",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_scale0p50_safe_types_margin0p20_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="counterfactual advantage abstention scale tie",
        pending_job="14862802/14862803",
    ),
    ResultSpec(
        key="retrieval_residual_knn2_scale040_safe_margin020",
        label="K2 train-state residual retrieval, scale 0.40, safe residuals, advantage margin 0.20",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_knn2_scale0p40_safe_types_margin0p20_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="previous best counterfactual advantage abstention",
        pending_job="14862936/14862937",
    ),
    ResultSpec(
        key="retrieval_residual_taskrelative_knn2_scale040_safe_margin020",
        label="K2 task-relative residual retrieval, scale 0.40, safe residuals, advantage margin 0.20",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_taskrelative_knn2_scale0p40_safe_types_margin0p20_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="task-relative state metric for counterfactual tangent retrieval",
        pending_job="14893789/14893790",
    ),
    ResultSpec(
        key="retrieval_residual_k1grid_tight_safe_ray_margin020",
        label="K1 train-state residual ray search, safe residuals, scales 0.30/0.40/0.50, advantage margin 0.20",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k1grid_tight_safe_ray_margin0p20_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="counterfactual tangent ray-search diagnostic",
        pending_job="14868993/14868994",
    ),
    ResultSpec(
        key="retrieval_residual_k2grid_tight_safe_ray_margin020",
        label="K2 train-state residual ray search, safe residuals, scales 0.30/0.40/0.50, advantage margin 0.20",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k2grid_tight_safe_ray_margin0p20_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="counterfactual tangent ray-search diagnostic",
        pending_job="14868995/14868996",
    ),
    ResultSpec(
        key="retrieval_residual_k2grid_broad_safe_ray_margin020",
        label="K2 train-state residual ray search, safe residuals, scales 0.20/0.35/0.50/0.65, advantage margin 0.20",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k2grid_broad_safe_ray_margin0p20_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="counterfactual tangent ray-search diagnostic",
        pending_job="14868997/14868998",
    ),
    ResultSpec(
        key="retrieval_residual_k4grid_tight_safe_ray_margin020",
        label="K4 train-state residual ray search, safe residuals, scales 0.30/0.40/0.50, advantage margin 0.20",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k4grid_tight_safe_ray_margin0p20_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="counterfactual tangent ray-search diagnostic",
        pending_job="14868999/14869000",
    ),
    ResultSpec(
        key="retrieval_residual_k4_scale040_safe_margin020_mean_by_type",
        label="K4 train-state residual retrieval, scale 0.40, safe residuals, mean-by-type tangent consensus",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k4s040_safe_margin0p20_mean_by_type_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="counterfactual tangent consensus near-tie ablation",
        pending_job="14868699/14868700",
    ),
    ResultSpec(
        key="retrieval_residual_k4_kernel_mean",
        label="K4 kernel-weighted residual retrieval, scale 0.40, margin 0.20",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k4s040_safe_margin0p20_kernel_mean_by_type_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="local counterfactual tangent-field interpolation",
        pending_job="14891067/14891083",
    ),
    ResultSpec(
        key="retrieval_residual_k4_kernel_mean_noopbonus003",
        label="K4 kernel-weighted residual retrieval, scale 0.40, margin 0.20, no-op residual bonus 0.03",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k4s040_safe_margin0p20_kernel_mean_by_type_noopbonus0p03_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="local counterfactual tangent-field interpolation with sparse-action prior",
        pending_job="14891072/14891085",
    ),
    ResultSpec(
        key="retrieval_residual_k4_kernel_mean_s035_noopbonus003",
        label="K4 kernel-weighted residual retrieval, scale 0.35, margin 0.20, no-op residual bonus 0.03",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k4s035_safe_margin0p20_kernel_mean_by_type_noopbonus0p03_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="local counterfactual tangent-field interpolation scale check",
        pending_job="14891076/14891087",
    ),
    ResultSpec(
        key="retrieval_residual_k4_kernel_mean_s045_noopbonus003",
        label="K4 kernel-weighted residual retrieval, scale 0.45, margin 0.20, no-op residual bonus 0.03",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k4s045_safe_margin0p20_kernel_mean_by_type_noopbonus0p03_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="local counterfactual tangent-field interpolation scale check",
        pending_job="14891082/14891088",
    ),
    ResultSpec(
        key="retrieval_residual_k4_fieldsoftmax_grid",
        label="K4 field-softmax residual transport, safe residuals, scales 0.35/0.40/0.45, margin 0.20",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k4_fieldsoftmax_grid_safe_margin0p20_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="field-conditioned counterfactual tangent transport",
        pending_job="14891889/14891934",
    ),
    ResultSpec(
        key="retrieval_residual_k4_fieldsoftmax_grid_noopbonus003",
        label="K4 field-softmax residual transport, safe residuals, scales 0.35/0.40/0.45, margin 0.20, no-op residual bonus 0.03",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k4_fieldsoftmax_grid_safe_margin0p20_noopbonus0p03_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="field-conditioned tangent transport with sparse-action prior",
        pending_job="14891902/14891946",
    ),
    ResultSpec(
        key="retrieval_residual_k4_fieldsoftmax_grid_margin010_noopbonus003",
        label="K4 field-softmax residual transport, safe residuals, scales 0.35/0.40/0.45, margin 0.10, no-op residual bonus 0.03",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k4_fieldsoftmax_grid_safe_margin0p10_noopbonus0p03_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="field-conditioned tangent transport abstention sweep",
        pending_job="14892958/14893002",
    ),
    ResultSpec(
        key="retrieval_residual_k4_fieldsoftmax_grid_margin005_noopbonus003",
        label="K4 field-softmax residual transport, safe residuals, scales 0.35/0.40/0.45, margin 0.05, no-op residual bonus 0.03",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k4_fieldsoftmax_grid_safe_margin0p05_noopbonus0p03_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="field-conditioned tangent transport abstention sweep",
        pending_job="14892975/14893016",
    ),
    ResultSpec(
        key="retrieval_residual_k4_fieldsoftmax_grid_margin000_noopbonus003",
        label="K4 field-softmax residual transport, safe residuals, scales 0.35/0.40/0.45, margin 0.00, no-op residual bonus 0.03",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k4_fieldsoftmax_grid_safe_margin0p00_noopbonus0p03_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="field-conditioned tangent transport no-abstention diagnostic",
        pending_job="14892990/14893028",
    ),
    ResultSpec(
        key="retrieval_residual_k8_fieldsoftmax_grid_noopbonus003",
        label="K8 field-softmax residual transport, safe residuals, scales 0.35/0.40/0.45, margin 0.20, no-op residual bonus 0.03",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k8_fieldsoftmax_grid_safe_margin0p20_noopbonus0p03_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="field-conditioned tangent transport neighborhood scaling",
        pending_job="14891923/14891960",
    ),
    ResultSpec(
        key="retrieval_residual_k4_mean_noopbonus003",
        label="K4 mean-by-type residual retrieval, scale 0.40, margin 0.20, no-op residual bonus 0.03",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k4s040_safe_margin0p20_mean_by_type_noopbonus0p03_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="current best clean typed sparse-intervention prior",
        pending_job="14883919/14883920",
    ),
    ResultSpec(
        key="retrieval_residual_k4_mean_noopbonus003_srcprog025",
        label="K4 mean-by-type residual retrieval, scale 0.40, margin 0.20, no-op bonus 0.03, source progress >= 0.25",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k4s040_safe_margin0p20_mean_by_type_noopbonus0p03_srcprog0p25_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="soft train-source viability gate for sparse residual transport",
        pending_job="14894438/14894439",
    ),
    ResultSpec(
        key="retrieval_residual_k4_mean_margin015_noopbonus003",
        label="K4 mean-by-type residual retrieval, scale 0.40, margin 0.15, no-op bonus 0.03",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k4s040_safe_margin0p15_mean_by_type_noopbonus0p03_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="advantage-abstention margin fine sweep for sparse residual transport",
        pending_job="14897841/14897845",
    ),
    ResultSpec(
        key="retrieval_residual_k4_mean_margin025_noopbonus003",
        label="K4 mean-by-type residual retrieval, scale 0.40, margin 0.25, no-op bonus 0.03",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k4s040_safe_margin0p25_mean_by_type_noopbonus0p03_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="advantage-abstention margin fine sweep for sparse residual transport",
        pending_job="14897842/14897846",
    ),
    ResultSpec(
        key="retrieval_residual_k4_mean_margin015_srcscorebonus002",
        label="K4 mean-by-type residual retrieval, scale 0.40, margin 0.15, source-score bonus 0.02",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k4s040_safe_margin0p15_mean_by_type_srcscorebonus0p02_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="advantage-abstention margin fine sweep with measured train-source prior",
        pending_job="14897843/14897847",
    ),
    ResultSpec(
        key="retrieval_residual_k4_mean_margin025_srcscorebonus002",
        label="K4 mean-by-type residual retrieval, scale 0.40, margin 0.25, source-score bonus 0.02",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k4s040_safe_margin0p25_mean_by_type_srcscorebonus0p02_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="advantage-abstention margin fine sweep with measured train-source prior",
        pending_job="14897844/14897848",
    ),
    ResultSpec(
        key="retrieval_residual_k4_mean_grid035040045_noopbonus003",
        label="K4 mean-by-type residual retrieval, scales 0.35/0.40/0.45, margin 0.20, no-op bonus 0.03",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k4_grid035040045_safe_margin0p20_mean_by_type_noopbonus0p03_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="scale-grid diagnostic for sparse mean-consensus residual transport",
        pending_job="14897988/14897990",
    ),
    ResultSpec(
        key="retrieval_residual_k4_mean_grid035040045_srcscorebonus002",
        label="K4 mean-by-type residual retrieval, scales 0.35/0.40/0.45, margin 0.20, source-score bonus 0.02",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k4_grid035040045_safe_margin0p20_mean_by_type_srcscorebonus0p02_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="scale-grid diagnostic for measured-prior mean-consensus residual transport",
        pending_job="14897989/14897991",
    ),
    ResultSpec(
        key="retrieval_residual_k4_mean_grid035040045_srcadvbonus002",
        label="K4 mean-by-type residual retrieval, scales 0.35/0.40/0.45, margin 0.20, source-advantage bonus 0.02",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k4_grid035040045_safe_margin0p20_mean_by_type_srcadvbonus0p02_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="source-local utility-lift prior for mean-consensus residual transport",
        pending_job="14902706/14902707",
    ),
    ResultSpec(
        key="retrieval_residual_k4_mean_grid035040045_srcadvbonus005",
        label="K4 mean-by-type residual retrieval, scales 0.35/0.40/0.45, margin 0.20, source-advantage bonus 0.05",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k4_grid035040045_safe_margin0p20_mean_by_type_srcadvbonus0p05_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="source-local utility-lift prior for mean-consensus residual transport",
        pending_job="14902709/14902711",
    ),
    ResultSpec(
        key="retrieval_residual_k4_mean_grid035040045_noopbonus003_srcadvbonus002",
        label="K4 mean-by-type residual retrieval, scales 0.35/0.40/0.45, margin 0.20, no-op bonus 0.03, source-advantage bonus 0.02",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k4_grid035040045_safe_margin0p20_mean_by_type_noopbonus0p03_srcadvbonus0p02_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="source-local utility-lift calibration on the current best typed prior",
        pending_job="14902713/14902715",
    ),
    ResultSpec(
        key="retrieval_residual_k4_mean_grid035040045_srcadvgate000",
        label="K4 mean-by-type residual retrieval, scales 0.35/0.40/0.45, margin 0.20, source-advantage gate >= 0.0",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k4_grid035040045_safe_margin0p20_mean_by_type_srcadvgate0p0_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="source-local utility-lift gate for mean-consensus residual transport",
        pending_job="14902717/14902719",
    ),
    ResultSpec(
        key="retrieval_residual_k4_mean_grid035040045_noopbonus003_srcadvgate000",
        label="K4 mean-by-type residual retrieval, scales 0.35/0.40/0.45, margin 0.20, no-op bonus 0.03, source-advantage gate >= 0.0",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k4_grid035040045_safe_margin0p20_mean_by_type_noopbonus0p03_srcadvgate0p0_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="source-local utility-lift gate on the current best typed prior",
        pending_job="14902721/14902723",
    ),
    ResultSpec(
        key="retrieval_residual_k4_mean_grid035040045_typesuccessbonus002",
        label="K4 mean-by-type residual retrieval, scales 0.35/0.40/0.45, margin 0.20, train family-success bonus 0.02",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k4_grid035040045_safe_margin0p20_mean_by_type_typesuccessbonus0p02_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="continuous train-family reliability prior for mean-consensus residual transport",
        pending_job="14903128/14903129",
    ),
    ResultSpec(
        key="retrieval_residual_k4_mean_grid035040045_typesuccessbonus003",
        label="K4 mean-by-type residual retrieval, scales 0.35/0.40/0.45, margin 0.20, train family-success bonus 0.03",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k4_grid035040045_safe_margin0p20_mean_by_type_typesuccessbonus0p03_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="continuous train-family reliability prior for mean-consensus residual transport",
        pending_job="14903130/14903131",
    ),
    ResultSpec(
        key="retrieval_residual_k4_mean_grid035040045_typesuccessbonus005",
        label="K4 mean-by-type residual retrieval, scales 0.35/0.40/0.45, margin 0.20, train family-success bonus 0.05",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k4_grid035040045_safe_margin0p20_mean_by_type_typesuccessbonus0p05_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="continuous train-family reliability prior for mean-consensus residual transport",
        pending_job="14903132/14903133",
    ),
    ResultSpec(
        key="retrieval_residual_k4_mean_grid035040045_noopbonus003_typesuccessbonus002",
        label="K4 mean-by-type residual retrieval, scales 0.35/0.40/0.45, margin 0.20, no-op bonus 0.03, train family-success bonus 0.02",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k4_grid035040045_safe_margin0p20_mean_by_type_noopbonus0p03_typesuccessbonus0p02_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="continuous train-family reliability calibration on the current best typed prior",
        pending_job="14903134/14903135",
    ),
    ResultSpec(
        key="retrieval_residual_k4_mean_grid035040045_consensus005",
        label="K4 mean-by-type residual retrieval, scales 0.35/0.40/0.45, margin 0.20, consensus penalty 0.05",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k4_grid035040045_safe_margin0p20_mean_by_type_consensus0p05_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="train-neighbor tangent-consensus confidence without sparse type prior",
        pending_job="14903384/14903385",
    ),
    ResultSpec(
        key="retrieval_residual_k4_mean_grid035040045_noopbonus003_consensus002",
        label="K4 mean-by-type residual retrieval, scales 0.35/0.40/0.45, margin 0.20, no-op bonus 0.03, consensus penalty 0.02",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k4_grid035040045_safe_margin0p20_mean_by_type_noopbonus0p03_consensus0p02_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="train-neighbor tangent-consensus confidence on the current best typed prior",
        pending_job="14903386/14903387",
    ),
    ResultSpec(
        key="retrieval_residual_k4_mean_grid035040045_noopbonus003_consensus005",
        label="K4 mean-by-type residual retrieval, scales 0.35/0.40/0.45, margin 0.20, no-op bonus 0.03, consensus penalty 0.05",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k4_grid035040045_safe_margin0p20_mean_by_type_noopbonus0p03_consensus0p05_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="train-neighbor tangent-consensus confidence on the current best typed prior",
        pending_job="14903388/14903389",
    ),
    ResultSpec(
        key="retrieval_residual_k4_mean_grid035040045_noopbonus003_consensus010",
        label="K4 mean-by-type residual retrieval, scales 0.35/0.40/0.45, margin 0.20, no-op bonus 0.03, consensus penalty 0.10",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k4_grid035040045_safe_margin0p20_mean_by_type_noopbonus0p03_consensus0p10_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="train-neighbor tangent-consensus confidence on the current best typed prior",
        pending_job="14903390/14903391",
    ),
    ResultSpec(
        key="retrieval_residual_k4_compose_grid035040045",
        label="K4 composed type-consensus residual retrieval, scales 0.35/0.40/0.45, margin 0.20",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k4_compose_grid035040045_safe_margin0p20_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="local tangent composition without typed priors",
        pending_job="14905956/14905963",
    ),
    ResultSpec(
        key="retrieval_residual_k4_compose_grid035040045_noopbonus003",
        label="K4 composed type-consensus residual retrieval, scales 0.35/0.40/0.45, margin 0.20, no-op bonus 0.03",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k4_compose_grid035040045_safe_margin0p20_noopbonus0p03_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="local tangent composition on the current best typed prior",
        pending_job="14905957/14905964",
    ),
    ResultSpec(
        key="retrieval_residual_k4_composemasked_grid035040045",
        label="K4 composed type-consensus residual retrieval, masked, scales 0.35/0.40/0.45, margin 0.20",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k4_composemasked_grid035040045_safe_margin0p20_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="local tangent composition with anti-goal composite masks",
        pending_job="14911979/14911982",
    ),
    ResultSpec(
        key="retrieval_residual_k4_composemasked_grid035040045_noopbonus003",
        label="K4 composed type-consensus residual retrieval, masked, scales 0.35/0.40/0.45, margin 0.20, no-op bonus 0.03",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k4_composemasked_grid035040045_safe_margin0p20_noopbonus0p03_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="local tangent composition with anti-goal composite masks on the current best typed prior",
        pending_job="14911980/14911983",
    ),
    ResultSpec(
        key="retrieval_residual_k4_composemasked_dropnmnoop_grid035040045",
        label="K4 composed type-consensus residual retrieval, masked, drop near-miss+no-op composite",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k4_composemasked_dropnmnoop_grid035040045_safe_margin0p20_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="exact compatibility mask without typed no-op prior",
        pending_job="14927919/14927921",
    ),
    ResultSpec(
        key="retrieval_residual_k4_composemasked_dropnmnoop_grid035040045_noopbonus003",
        label="K4 composed type-consensus residual retrieval, masked, drop near-miss+no-op composite",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k4_composemasked_dropnmnoop_grid035040045_safe_margin0p20_noopbonus0p03_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="exact incompatibility mask for one weak composed tangent pair",
        pending_job="14915009/14915013",
    ),
    ResultSpec(
        key="retrieval_residual_k4_composemasked_dropnmnoop_grid035040045_margin010_noopbonus003",
        label="K4 compatible residual retrieval, margin 0.10, no-op bonus 0.03",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k4_composemasked_dropnmnoop_grid035040045_safe_margin0p10_noopbonus0p03_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="oracle-motivated abstention sensitivity on the compatible local tangent chart",
        pending_job="14936131/14936132",
    ),
    ResultSpec(
        key="retrieval_residual_k4_composemasked_dropnmnoop_grid035040045_noopbonus003_oraclek8",
        label="K4 compatible residual retrieval, unique candidate-oracle prefix K=8 diagnostic",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k4_composemasked_dropnmnoop_grid035040045_safe_margin0p20_noopbonus0p03_oraclek8_summary.json",
        clean_deployment="diagnostic",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="diagnostic oracle over generated clean candidate prefix to separate proposal headroom from field ranking",
        pending_job="14953513/14953522",
    ),
    ResultSpec(
        key="retrieval_residual_k4_composemasked_dropnmnoop_grid035040045_noopbonus003_oraclek8trace",
        label="K4 compatible residual retrieval, unique candidate-oracle prefix K=8 branch trace",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k4_composemasked_dropnmnoop_grid035040045_safe_margin0p20_noopbonus0p03_oraclek8trace_summary.json",
        clean_deployment="diagnostic",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="diagnostic branch trace for deployable selector calibration over the local tangent chart",
        pending_job="14953960/14953961",
    ),
    ResultSpec(
        key="retrieval_residual_k4_composemasked_dropnmnoop_grid035040045_noopbonus003_nmchallenger002",
        label="K4 compatible residual retrieval, near-miss challenger gate 0.02",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k4_composemasked_dropnmnoop_grid035040045_safe_margin0p20_noopbonus0p03_nmchallenger0p02_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="trace-motivated two-stage selector calibration over the compatible local tangent chart",
        pending_job="14954280/14954281",
    ),
    ResultSpec(
        key="retrieval_residual_k4_composemasked_dropnmnoop_grid035040045_noopbonus003_nmchallenger001",
        label="K4 compatible residual retrieval, near-miss challenger gate 0.01",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k4_composemasked_dropnmnoop_grid035040045_safe_margin0p20_noopbonus0p03_nmchallenger0p01_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="lower-margin sensitivity for trace-motivated singleton near-miss challenger calibration",
        pending_job="14954526/14954527",
    ),
    ResultSpec(
        key="retrieval_residual_k4_composemasked_dropnmnoop_grid035040045_noopbonus003_nmchallenger001_stacknowg",
        label="K4 compatible residual retrieval, near-miss challenger gate 0.01, no wrong-gripper component on Stack",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k4_composemasked_dropnmnoop_grid035040045_safe_margin0p20_noopbonus0p03_nmchallenger0p01_stacknowg_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="task-local tangent compatibility mask for Stack where wrong-gripper residuals are an anti-affordance",
        pending_job="14976136/14976144",
    ),
    ResultSpec(
        key="retrieval_residual_k4_composemasked_dropnmnoop_grid035040045_noopbonus003_nmchallenger0005",
        label="K4 compatible residual retrieval, near-miss challenger gate 0.005",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k4_composemasked_dropnmnoop_grid035040045_safe_margin0p20_noopbonus0p03_nmchallenger0p005_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="fine lower-margin calibration for the trace-motivated singleton near-miss challenger gate",
        pending_job="14955089/14955090",
    ),
    ResultSpec(
        key="retrieval_residual_k4_composemasked_dropnmnoop_grid035040045_noopbonus003_nmchallenger0015",
        label="K4 compatible residual retrieval, near-miss challenger gate 0.015",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k4_composemasked_dropnmnoop_grid035040045_safe_margin0p20_noopbonus0p03_nmchallenger0p015_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="fine upper-margin calibration for the trace-motivated singleton near-miss challenger gate",
        pending_job="14955093/14955094",
    ),
    ResultSpec(
        key="retrieval_residual_k4_composemasked_dropnmnoop_grid035040045_noopbonus003_nmchallenger001_scale035",
        label="K4 compatible residual retrieval, near-miss challenger gate 0.01, scale-gated 0.35",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k4_composemasked_dropnmnoop_grid035040045_safe_margin0p20_noopbonus0p03_nmchallenger0p01_scale035_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="trace-motivated near-miss challenger with tangent-length reliability gating at the shortest scale",
        pending_job="14955663/14955664",
    ),
    ResultSpec(
        key="retrieval_residual_k4_composemasked_dropnmnoop_grid035040045_noopbonus003_nmchallenger001_scales035040",
        label="K4 compatible residual retrieval, near-miss challenger gate 0.01, scale-gated 0.35/0.40",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k4_composemasked_dropnmnoop_grid035040045_safe_margin0p20_noopbonus0p03_nmchallenger0p01_scales035040_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="trace-motivated near-miss challenger with tangent-length reliability gating over the two shortest scales",
        pending_job="14955665/14955666",
    ),
    ResultSpec(
        key="retrieval_residual_k4_composemasked_dropnmnoop_grid035040045_noopbonus003_nmchallenger003",
        label="K4 compatible residual retrieval, near-miss challenger gate 0.03",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k4_composemasked_dropnmnoop_grid035040045_safe_margin0p20_noopbonus0p03_nmchallenger0p03_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="upper-margin sensitivity for trace-motivated singleton near-miss challenger calibration",
        pending_job="14954528/14954529",
    ),
    ResultSpec(
        key="retrieval_residual_k4_composemasked_dropnmnoop_grid035040045_noopbonus003_nmwgchallenger001",
        label="K4 compatible residual retrieval, near-miss/wrong-gripper challenger gate 0.01",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k4_composemasked_dropnmnoop_grid035040045_safe_margin0p20_noopbonus0p03_nmwgchallenger0p01_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="trace-motivated challenger calibration test for whether wrong-gripper residuals carry conditional selector headroom",
        pending_job="14954530/14954531",
    ),
    ResultSpec(
        key="retrieval_residual_k4_composemasked_dropnmnoop_grid035040045_noopbonus003_nmwgmargin003_challenger001",
        label="K4 compatible residual retrieval, near-miss challenger 0.01, wrong-gripper margin 0.03",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k4_composemasked_dropnmnoop_grid035040045_safe_margin0p20_noopbonus0p03_nmwgmargin0p03_challenger0p01_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="family-specific reliability margin: keep near-miss sensitive while requiring stronger wrong-gripper evidence",
        pending_job="14989648/14989649",
    ),
    ResultSpec(
        key="retrieval_residual_k4_composemasked_dropnmnoop_grid035040045_noopbonus003_nmwgmargin005_challenger001",
        label="K4 compatible residual retrieval, near-miss challenger 0.01, wrong-gripper margin 0.05",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k4_composemasked_dropnmnoop_grid035040045_safe_margin0p20_noopbonus0p03_nmwgmargin0p05_challenger0p01_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="family-specific reliability margin upper setting for wrong-gripper challenger evidence",
        pending_job="14989651/14989659",
    ),
    ResultSpec(
        key="retrieval_residual_k4_composemasked_dropnmnoop_grid035040045_noopbonus003_nmwgchallenger001_pickpull",
        label="K4 compatible residual retrieval, near-miss/wrong-gripper challenger gate 0.01 on Pick/Pull only",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k4_composemasked_dropnmnoop_grid035040045_safe_margin0p20_noopbonus0p03_nmwgchallenger0p01_pickpull_norm_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="task-conditional challenger gate using branch trace evidence that wrong-gripper tangents help Pick/Pull but harm other tasks",
        pending_job="14974602/14974608",
    ),
    ResultSpec(
        key="retrieval_residual_k4_composemasked_dropnmnoop_grid035040045_noopbonus003_nmglobal_wgpickpull_challenger001",
        label="K4 compatible residual retrieval, near-miss global + wrong-gripper challenger on Pick/Pull, gate 0.01",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k4_composemasked_dropnmnoop_grid035040045_safe_margin0p20_noopbonus0p03_nmglobal_wgpickpull_challenger0p01_norm_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="type-specific affordance gate: keep near-miss global while restricting wrong-gripper challenger to acquisition/contact tasks",
        pending_job="14974610/14974611",
    ),
    ResultSpec(
        key="retrieval_residual_k4_composemasked_dropnmnoop_grid035040045_noopbonus003_nmglobal_wgpickpull_wgmargin003_challenger001",
        label="K4 compatible residual retrieval, near-miss global + wrong-gripper Pick/Pull margin 0.03",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k4_composemasked_dropnmnoop_grid035040045_safe_margin0p20_noopbonus0p03_nmglobal_wgpickpull_wgmargin0p03_challenger0p01_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="type-specific affordance margin: keep near-miss globally sensitive while requiring stronger wrong-gripper evidence only on Pick/Pull",
        pending_job="14991117/14991125",
    ),
    ResultSpec(
        key="retrieval_residual_k4_composemasked_dropnmnoop_grid035040045_noopbonus003_nmglobal_wgpickpull_wgmargin005_challenger001",
        label="K4 compatible residual retrieval, near-miss global + wrong-gripper Pick/Pull margin 0.05",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k4_composemasked_dropnmnoop_grid035040045_safe_margin0p20_noopbonus0p03_nmglobal_wgpickpull_wgmargin0p05_challenger0p01_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="upper setting for type-specific wrong-gripper evidence on Pick/Pull while preserving global near-miss calibration",
        pending_job="14991123/14991126",
    ),
    ResultSpec(
        key="retrieval_residual_k4_composemasked_dropnmnoop_grid035040045_noopbonus003_nmglobal_wgpickpullstack_challenger001",
        label="K4 compatible residual retrieval, near-miss global + wrong-gripper challenger on Pick/Pull/Stack",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k4_composemasked_dropnmnoop_grid035040045_safe_margin0p20_noopbonus0p03_nmglobal_wgpickpullstack_challenger0p01_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="oracle-trace contact-task calibration: wrong-gripper residuals are only challenger-eligible on Pick/Pull/Stack while near-miss stays global",
        pending_job="14995028/14995047",
    ),
    ResultSpec(
        key="retrieval_residual_k4_composemasked_dropnmnoop_grid035040045_noopbonus003_nmglobal_noopwgcontact_challenger001",
        label="K4 compatible residual retrieval, near-miss global + no-op/wrong-gripper contact challenger",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k4_composemasked_dropnmnoop_grid035040045_safe_margin0p20_noopbonus0p03_nmglobal_noopwgcontact_challenger0p01_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="contact-task sparse-family challenger ablation using no-op and wrong-gripper residuals where oracle trace shows proposal headroom",
        pending_job="14995045/14995048",
    ),
    ResultSpec(
        key="retrieval_residual_k6_composemasked_dropnmnoop_grid035040045_noopbonus003_nmchallenger001",
        label="K6 compatible residual retrieval, near-miss challenger gate 0.01",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k6_composemasked_dropnmnoop_grid035040045_safe_margin0p20_noopbonus0p03_nmchallenger0p01_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="neighborhood-size test for whether compatible transported tangent charts need more train-state support than K4",
        pending_job="14995739/14995759",
    ),
    ResultSpec(
        key="retrieval_residual_k8_composemasked_dropnmnoop_grid035040045_noopbonus003_nmchallenger001",
        label="K8 compatible residual retrieval, near-miss challenger gate 0.01",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k8_composemasked_dropnmnoop_grid035040045_safe_margin0p20_noopbonus0p03_nmchallenger0p01_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="larger-neighborhood transported tangent support test after K4 plateaued near 36%",
        pending_job="14995741/14995761",
    ),
    ResultSpec(
        key="retrieval_residual_k4_composemasked_dropnmnoop_grid035040045_noopbonus003_typesuccessbonus002_nmchallenger001",
        label="K4 compatible residual retrieval, train family-success bonus 0.02, near-miss challenger gate 0.01",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k4_composemasked_dropnmnoop_grid035040045_safe_margin0p20_noopbonus0p03_typesuccessbonus0p02_nmchallenger0p01_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="train-split counterfactual family-reliability prior on the current compatible tangent chart",
        pending_job="14995944/14995979",
    ),
    ResultSpec(
        key="retrieval_residual_k4_composemasked_dropnmnoop_grid035040045_noopbonus003_typesuccessbonus005_nmchallenger001",
        label="K4 compatible residual retrieval, train family-success bonus 0.05, near-miss challenger gate 0.01",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k4_composemasked_dropnmnoop_grid035040045_safe_margin0p20_noopbonus0p03_typesuccessbonus0p05_nmchallenger0p01_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="stronger train-family reliability prior on the current compatible tangent chart",
        pending_job="14995951/14995991",
    ),
    ResultSpec(
        key="typed_proposal_lattice_types6_prepend_margin000",
        label="Typed proposal lattice head, six families, policy-prepended margin 0.00",
        path="h16_policy_ckpt_near_miss_policy_bc5_typedprop_p2_bestpt_proposal_lattice_types6_prepend_margin0p00_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="model-generated counterfactual proposal support test; field scores typed proposals rather than retrieved train-state residuals",
        pending_job="14962264/14962356/14962363",
    ),
    ResultSpec(
        key="typed_proposal_lattice_types6_prepend_margin005",
        label="Typed proposal lattice head, six families, policy-prepended margin 0.05",
        path="h16_policy_ckpt_near_miss_policy_bc5_typedprop_p2_bestpt_proposal_lattice_types6_prepend_margin0p05_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="model-generated counterfactual proposal support test with a conservative policy-abstention margin",
        pending_job="14962264/14962357/14962364",
    ),
    ResultSpec(
        key="typed_proposal_lattice_types4safe_prepend_margin005",
        label="Typed proposal lattice head, safe four families, policy-prepended margin 0.05",
        path="h16_policy_ckpt_near_miss_policy_bc5_typedprop_p2_bestpt_proposal_lattice_types4safe_prepend_margin0p05_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="anti-goal proposal-family mask test for the model-generated typed lattice",
        pending_job="14963030/14963044",
    ),
    ResultSpec(
        key="typed_proposal_lattice_types2sparse_prepend_margin005",
        label="Typed proposal lattice head, sparse no-op/wrong-gripper families, policy-prepended margin 0.05",
        path="h16_policy_ckpt_near_miss_policy_bc5_typedprop_p2_bestpt_proposal_lattice_types2sparse_prepend_margin0p05_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="sparse safe-support test for whether typed generated no-op/wrong-gripper proposals recover the clean support gap",
        pending_job="14963031/14963042",
    ),
    ResultSpec(
        key="typed_proposal_lattice_nooponly_prepend_margin005",
        label="Typed proposal lattice head, no-op family only, policy-prepended margin 0.05",
        path="h16_policy_ckpt_near_miss_policy_bc5_typedprop_p2_bestpt_proposal_lattice_nooponly_prepend_margin0p05_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="minimal generated no-op proposal-family test after sparse typed lattice improved over the full six-family row",
        pending_job="14963032/14963043",
    ),
    ResultSpec(
        key="typed_proposal_lattice_types2sparse_bestpolicy_prepend_margin005",
        label="Typed proposal lattice head, sparse no-op/wrong-gripper families, best-policy checkpoint",
        path="h16_bestpolicy_types2sparse_prepend_margin0p05_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="checkpointing ablation for typed generated sparse proposals: lower BC/proposal loss versus weaker field scorer",
        pending_job="14963143/14963150",
    ),
    ResultSpec(
        key="typed_proposal_lattice_nooponly_bestpolicy_prepend_margin005",
        label="Typed proposal lattice head, no-op family only, best-policy checkpoint",
        path="h16_bestpolicy_nooponly_prepend_margin0p05_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="minimal generated no-op checkpointing ablation for typed proposal support",
        pending_job="14963147/14963149",
    ),
    ResultSpec(
        key="typed_proposal_lattice_nooponly_sparsehead_prepend_margin005",
        label="Typed proposal lattice head, no-op-only head retrain",
        path="h16_typedprop_nooponly_sparsehead_prepend_margin0p05_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="sparse-head retrain test for whether typed proposal failure comes from multi-family interference",
        pending_job="14963500/14963505/14963507",
    ),
    ResultSpec(
        key="typed_proposal_lattice_nooponly_sparsehead_prepend_margin000",
        label="Typed proposal lattice head, no-op-only head retrain, margin 0.00",
        path="h16_typedprop_nooponly_sparsehead_prepend_margin0p00_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="abstention calibration for the no-op-only typed proposal head: select generated no-op more often",
        pending_job="14964011/14964019",
    ),
    ResultSpec(
        key="typed_proposal_lattice_nooponly_sparsehead_prepend_margin010",
        label="Typed proposal lattice head, no-op-only head retrain, margin 0.10",
        path="h16_typedprop_nooponly_sparsehead_prepend_margin0p10_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="abstention calibration for the no-op-only typed proposal head: require stronger field advantage",
        pending_job="14964016/14964020",
    ),
    ResultSpec(
        key="retrieval_residual_k4_composemasked_dropnmnoop_grid035040045_noopbonus003_nmbonus001",
        label="K4 composed compatible residual retrieval, no-op bonus 0.03, singleton near-miss bonus 0.01",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k4_composemasked_dropnmnoop_grid035040045_safe_margin0p20_noopbonus0p03_nmbonus0p01_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="revive high-precision singleton near-miss tangents without boosting toxic near-miss+no-op composites",
        pending_job="14929087/14929152",
    ),
    ResultSpec(
        key="retrieval_residual_k4_composemasked_dropnmnoop_grid035040045_noopbonus003_nmbonus002",
        label="K4 composed compatible residual retrieval, no-op bonus 0.03, singleton near-miss bonus 0.02",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k4_composemasked_dropnmnoop_grid035040045_safe_margin0p20_noopbonus0p03_nmbonus0p02_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="stronger singleton near-miss revival prior on the compatible local tangent chart",
        pending_job="14929120/14929175",
    ),
    ResultSpec(
        key="retrieval_residual_k4_composemasked_dropnmnoop_grid035040045_srcscorebonus002",
        label="K4 composed type-consensus residual retrieval, masked, drop near-miss+no-op composite, source-score bonus 0.02",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k4_composemasked_dropnmnoop_grid035040045_safe_margin0p20_srcscorebonus0p02_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="train-measured source-score prior on the compatible local tangent chart",
        pending_job="14926713/14926723",
    ),
    ResultSpec(
        key="retrieval_residual_k4_composemasked_dropnmnoop_grid035040045_noopbonus003_srcscorebonus002",
        label="K4 composed type-consensus residual retrieval, masked, drop near-miss+no-op composite, no-op bonus 0.03, source-score bonus 0.02",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k4_composemasked_dropnmnoop_grid035040045_safe_margin0p20_noopbonus0p03_srcscorebonus0p02_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="train-measured source-score calibration on the current compatible typed-prior row",
        pending_job="14926714/14926726",
    ),
    ResultSpec(
        key="retrieval_residual_k4_composemasked_dropnmnoop_l2comp002_grid035040045_noopbonus003",
        label="K4 composed type-consensus residual retrieval, masked, drop near-miss+no-op composite, composite L2 penalty 0.02",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k4_composemasked_dropnmnoop_l2comp002_grid035040045_safe_margin0p20_noopbonus0p03_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="compatible local tangent composition with trust-radius regularization",
        pending_job="14915213/14915217",
    ),
    ResultSpec(
        key="retrieval_residual_k4_composemasked_compbonus_grid035040045_noopbonus003",
        label="K4 composed type-consensus residual retrieval, masked, component no-op bonus 0.03",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k4_composemasked_compbonus_grid035040045_safe_margin0p20_noopbonus0p03_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="component-wise sparse prior on the masked local tangent composition chart",
        pending_job="14912561/14912562",
    ),
    ResultSpec(
        key="retrieval_residual_k4_composemasked_l2comp002_grid035040045_noopbonus003",
        label="K4 composed type-consensus residual retrieval, masked, composite L2 penalty 0.02",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k4_composemasked_l2comp002_grid035040045_safe_margin0p20_noopbonus0p03_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="trust-radius penalty on composed local tangent candidates",
        pending_job="14913944/14913955",
    ),
    ResultSpec(
        key="retrieval_residual_k4_composemasked_l2comp005_grid035040045_noopbonus003",
        label="K4 composed type-consensus residual retrieval, masked, composite L2 penalty 0.05",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k4_composemasked_l2comp005_grid035040045_safe_margin0p20_noopbonus0p03_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="stronger trust-radius penalty on composed local tangent candidates",
        pending_job="14913951/14913956",
    ),
    ResultSpec(
        key="retrieval_repair_nearmiss_k4_grid025035050_margin020",
        label="K4 near-miss-to-expert repair tangent, scales 0.25/0.35/0.50, margin 0.20",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_repair_nearmiss_k4_grid025035050_margin0p20_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="deployment-clean corrective tangent transport from train near-misses back toward expert actions",
        pending_job="14904737/14904738",
    ),
    ResultSpec(
        key="retrieval_repair_nearmiss_k4_grid035050075_margin020",
        label="K4 near-miss-to-expert repair tangent, scales 0.35/0.50/0.75, margin 0.20",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_repair_nearmiss_k4_grid035050075_margin0p20_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="repair-tangent scale diagnostic for near-miss counterfactual geometry",
        pending_job="14904740/14904741",
    ),
    ResultSpec(
        key="retrieval_repair_nearmiss_k4_grid025035050_margin010",
        label="K4 near-miss-to-expert repair tangent, scales 0.25/0.35/0.50, margin 0.10",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_repair_nearmiss_k4_grid025035050_margin0p10_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="repair-tangent abstention diagnostic for near-miss counterfactual geometry",
        pending_job="14904742/14904743",
    ),
    ResultSpec(
        key="retrieval_repair_safe_k4_grid025035050_margin020",
        label="K4 safe-family-to-expert repair tangent, scales 0.25/0.35/0.50, margin 0.20",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_repair_safe_k4_grid025035050_margin0p20_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="repair-tangent family diagnostic including near-miss, no-op, and wrong-gripper corrections",
        pending_job="14904744/14904745",
    ),
    ResultSpec(
        key="retrieval_residual_k4_mean_grid035040045_noopbonus003_l2penalty005",
        label="K4 mean-by-type residual retrieval, scales 0.35/0.40/0.45, margin 0.20, no-op bonus 0.03, action L2 penalty 0.05",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k4_grid035040045_safe_margin0p20_mean_by_type_noopbonus0p03_l2penalty0p05_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="minimum-energy tangent diagnostic for sparse mean-consensus residual transport",
        pending_job="14898327/14898328",
    ),
    ResultSpec(
        key="retrieval_residual_k4_mean_grid035040045_noopbonus003_l2penalty010",
        label="K4 mean-by-type residual retrieval, scales 0.35/0.40/0.45, margin 0.20, no-op bonus 0.03, action L2 penalty 0.10",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k4_grid035040045_safe_margin0p20_mean_by_type_noopbonus0p03_l2penalty0p10_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="minimum-energy tangent diagnostic for sparse mean-consensus residual transport",
        pending_job="14898329/14898330",
    ),
    ResultSpec(
        key="retrieval_residual_k4_mean_grid035040045_noopbonus003_l2penalty020",
        label="K4 mean-by-type residual retrieval, scales 0.35/0.40/0.45, margin 0.20, no-op bonus 0.03, action L2 penalty 0.20",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k4_grid035040045_safe_margin0p20_mean_by_type_noopbonus0p03_l2penalty0p20_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="minimum-energy tangent diagnostic for sparse mean-consensus residual transport",
        pending_job="14898331/14898332",
    ),
    ResultSpec(
        key="retrieval_residual_k4_mean_grid040045050_noopbonus003",
        label="K4 mean-by-type residual retrieval, scales 0.40/0.45/0.50, margin 0.20, no-op bonus 0.03",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k4_grid040045050_safe_margin0p20_mean_by_type_noopbonus0p03_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="upper tangent-length sweep for sparse mean-consensus residual transport",
        pending_job="14898107/14898110",
    ),
    ResultSpec(
        key="retrieval_residual_k4_mean_grid040045050_srcscorebonus002",
        label="K4 mean-by-type residual retrieval, scales 0.40/0.45/0.50, margin 0.20, source-score bonus 0.02",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k4_grid040045050_safe_margin0p20_mean_by_type_srcscorebonus0p02_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="upper tangent-length sweep for measured-prior mean-consensus residual transport",
        pending_job="14898108/14898111",
    ),
    ResultSpec(
        key="retrieval_residual_k4_mean_grid035045055_noopbonus003",
        label="K4 mean-by-type residual retrieval, scales 0.35/0.45/0.55, margin 0.20, no-op bonus 0.03",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k4_grid035045055_safe_margin0p20_mean_by_type_noopbonus0p03_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="wide tangent-length sweep for sparse mean-consensus residual transport",
        pending_job="14898109/14898112",
    ),
    ResultSpec(
        key="retrieval_residual_k4_mean_nooponly_noopbonus003",
        label="K4 mean-by-type residual retrieval, scale 0.40, margin 0.20, no-op-only residuals, no-op bonus 0.03",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k4s040_safe_margin0p20_mean_by_type_nooponly_noopbonus0p03_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="no-op-only residual-family ablation for sparse tangent transport",
        pending_job="14897563/14897565",
    ),
    ResultSpec(
        key="retrieval_residual_k4_mean_nooponly_srcscorebonus002",
        label="K4 mean-by-type residual retrieval, scale 0.40, margin 0.20, no-op-only residuals, source-score bonus 0.02",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k4s040_safe_margin0p20_mean_by_type_nooponly_srcscorebonus0p02_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="no-op-only residual-family ablation with measured train-source prior",
        pending_job="14897564/14897566",
    ),
    ResultSpec(
        key="retrieval_residual_k4_mean_noopbonus003_srcprog050",
        label="K4 mean-by-type residual retrieval, scale 0.40, margin 0.20, no-op bonus 0.03, source progress >= 0.50",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k4s040_safe_margin0p20_mean_by_type_noopbonus0p03_srcprog0p50_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="train-source viability gate for sparse residual transport",
        pending_job="14894298/14894300",
    ),
    ResultSpec(
        key="retrieval_residual_k4_mean_noopbonus003_srcprog075",
        label="K4 mean-by-type residual retrieval, scale 0.40, margin 0.20, no-op bonus 0.03, source progress >= 0.75",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k4s040_safe_margin0p20_mean_by_type_noopbonus0p03_srcprog0p75_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="strict train-source viability gate for sparse residual transport",
        pending_job="14894299/14894301",
    ),
    ResultSpec(
        key="retrieval_residual_k4_mean_srcprogbonus003",
        label="K4 mean-by-type residual retrieval, scale 0.40, margin 0.20, source-progress bonus 0.03",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k4s040_safe_margin0p20_mean_by_type_srcprogbonus0p03_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="train-source progress prior replacing typed no-op prior",
        pending_job="14894674/14894676",
    ),
    ResultSpec(
        key="retrieval_residual_k4_mean_srcprogbonus005",
        label="K4 mean-by-type residual retrieval, scale 0.40, margin 0.20, source-progress bonus 0.05",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k4s040_safe_margin0p20_mean_by_type_srcprogbonus0p05_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="strong train-source progress prior replacing typed no-op prior",
        pending_job="14894675/14894677",
    ),
    ResultSpec(
        key="retrieval_residual_k4_mean_srcscorebonus0015",
        label="K4 mean-by-type residual retrieval, scale 0.40, margin 0.20, source-score bonus 0.015",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k4s040_safe_margin0p20_mean_by_type_srcscorebonus0p015_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="train-source reward-score prior for sparse residual transport",
        pending_job="14897123/14897126",
    ),
    ResultSpec(
        key="retrieval_residual_k4_mean_srcscorebonus002",
        label="K4 mean-by-type residual retrieval, scale 0.40, margin 0.20, source-score bonus 0.02",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k4s040_safe_margin0p20_mean_by_type_srcscorebonus0p02_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="train-source reward-score prior for sparse residual transport",
        pending_job="14897124/14897127",
    ),
    ResultSpec(
        key="retrieval_residual_k4_mean_srcscorebonus0025",
        label="K4 mean-by-type residual retrieval, scale 0.40, margin 0.20, source-score bonus 0.025",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k4s040_safe_margin0p20_mean_by_type_srcscorebonus0p025_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="train-source reward-score prior for sparse residual transport",
        pending_job="14897125/14897128",
    ),
    ResultSpec(
        key="retrieval_residual_taskrelative_k4_mean_noopbonus003",
        label="K4 task-relative mean-by-type residual retrieval, scale 0.40, margin 0.20, no-op residual bonus 0.03",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_taskrelative_k4s040_safe_margin0p20_mean_by_type_noopbonus0p03_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="task-relative state metric for typed sparse-intervention prior",
        pending_job="14893787/14893788",
    ),
    ResultSpec(
        key="retrieval_residual_k4_mean_noopbonus001",
        label="K4 mean-by-type residual retrieval, scale 0.40, margin 0.20, no-op residual bonus 0.01",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k4s040_safe_margin0p20_mean_by_type_noopbonus0p01_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="typed sparse-intervention prior fine sweep",
        pending_job="14884375/14884376",
    ),
    ResultSpec(
        key="retrieval_residual_k4_mean_noopbonus002",
        label="K4 mean-by-type residual retrieval, scale 0.40, margin 0.20, no-op residual bonus 0.02",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k4s040_safe_margin0p20_mean_by_type_noopbonus0p02_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="typed sparse-intervention prior fine sweep",
        pending_job="14884377/14884378",
    ),
    ResultSpec(
        key="retrieval_residual_k4_mean_noopbonus0025",
        label="K4 mean-by-type residual retrieval, scale 0.40, margin 0.20, no-op residual bonus 0.025",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k4s040_safe_margin0p20_mean_by_type_noopbonus0p025_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="typed sparse-intervention prior fine sweep",
        pending_job="14884379/14884380",
    ),
    ResultSpec(
        key="retrieval_residual_k4_mean_noopbonus0035",
        label="K4 mean-by-type residual retrieval, scale 0.40, margin 0.20, no-op residual bonus 0.035",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k4s040_safe_margin0p20_mean_by_type_noopbonus0p035_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="typed sparse-intervention prior fine sweep",
        pending_job="14884381/14884382",
    ),
    ResultSpec(
        key="retrieval_residual_k4_mean_wgbonus003",
        label="K4 mean-by-type residual retrieval, scale 0.40, margin 0.20, wrong-gripper residual bonus 0.03",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k4s040_safe_margin0p20_mean_by_type_wgbonus0p03_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="wrong-gripper typed sparse-intervention prior diagnostic",
        pending_job="14890071/14890072",
    ),
    ResultSpec(
        key="retrieval_residual_k4_mean_noop003_wg002",
        label="K4 mean-by-type residual retrieval, scale 0.40, margin 0.20, no-op 0.03 + wrong-gripper 0.02",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k4s040_safe_margin0p20_mean_by_type_noop0p03_wg0p02_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="two-family typed sparse-intervention prior diagnostic",
        pending_job="14890073/14890074",
    ),
    ResultSpec(
        key="retrieval_residual_k4_mean_noop003_wg004",
        label="K4 mean-by-type residual retrieval, scale 0.40, margin 0.20, no-op 0.03 + wrong-gripper 0.04",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k4s040_safe_margin0p20_mean_by_type_noop0p03_wg0p04_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="two-family typed sparse-intervention prior diagnostic",
        pending_job="14890075/14890076",
    ),
    ResultSpec(
        key="retrieval_residual_k4_mean_noop0025_wg002",
        label="K4 mean-by-type residual retrieval, scale 0.40, margin 0.20, no-op 0.025 + wrong-gripper 0.02",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k4s040_safe_margin0p20_mean_by_type_noop0p025_wg0p02_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="two-family typed sparse-intervention prior diagnostic",
        pending_job="14890077/14890078",
    ),
    ResultSpec(
        key="retrieval_residual_k4_mean_noopbonus005",
        label="K4 mean-by-type residual retrieval, scale 0.40, margin 0.20, no-op residual bonus 0.05",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k4s040_safe_margin0p20_mean_by_type_noopbonus0p05_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="typed sparse-intervention prior diagnostic",
        pending_job="14883921/14883922",
    ),
    ResultSpec(
        key="retrieval_residual_k4_mean_noopbonus008",
        label="K4 mean-by-type residual retrieval, scale 0.40, margin 0.20, no-op residual bonus 0.08",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_k4s040_safe_margin0p20_mean_by_type_noopbonus0p08_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="typed sparse-intervention prior diagnostic",
        pending_job="14883923/14883924",
    ),
    ResultSpec(
        key="retrieval_residual_policy_anchor_scale035_safe",
        label="Policy-relative train-state residual retrieval, scale 0.35, safe non-expert residuals",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_policy_anchor_scale0p35_safe_noexpert_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="policy-relative tangent anchor diagnostic",
        pending_job="14862605/14862606",
    ),
    ResultSpec(
        key="retrieval_residual_scale030_safe_types",
        label="Train-state residual retrieval, scale 0.30, policy/no-op/wrong-gripper residuals",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_scale0p30_safe_types_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="typed tangent scale zoom sweep",
        pending_job="14859589/14859590",
    ),
    ResultSpec(
        key="retrieval_residual_scale0325_safe_types",
        label="Train-state residual retrieval, scale 0.325, policy/no-op/wrong-gripper residuals",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_scale0p325_safe_types_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="typed tangent scale zoom sweep",
        pending_job="14859591/14859592",
    ),
    ResultSpec(
        key="retrieval_residual_scale0375_safe_types",
        label="Train-state residual retrieval, scale 0.375, policy/no-op/wrong-gripper residuals",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_scale0p375_safe_types_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="typed tangent scale zoom sweep",
        pending_job="14859593/14859594",
    ),
    ResultSpec(
        key="retrieval_residual_scale040_safe_types",
        label="Train-state residual retrieval, scale 0.40, policy/no-op/wrong-gripper residuals",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_scale0p40_safe_types_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="typed tangent scale zoom sweep",
        pending_job="14859595/14859596",
    ),
    ResultSpec(
        key="retrieval_residual_scale045_safe_types",
        label="Train-state residual retrieval, scale 0.45, policy/no-op/wrong-gripper residuals",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_scale0p45_safe_types_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="typed tangent scale fine sweep",
        pending_job="14859505/14859506",
    ),
    ResultSpec(
        key="retrieval_residual_scale060_safe_types",
        label="Train-state residual retrieval, scale 0.60, policy/no-op/wrong-gripper residuals",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_scale0p60_safe_types_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="typed tangent scale fine sweep",
        pending_job="14859507/14859508",
    ),
    ResultSpec(
        key="retrieval_residual_scale070_safe_types",
        label="Train-state residual retrieval, scale 0.70, policy/no-op/wrong-gripper residuals",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_scale0p70_safe_types_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="typed tangent scale fine sweep",
        pending_job="14859509/14859510",
    ),
    ResultSpec(
        key="retrieval_residual_scale050_type_success010",
        label="Train-state residual retrieval, scale 0.50, train family success >= 0.10",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_scale0p50_type_success0p10_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="train-split residual family reliability prior",
        pending_job="14859293/14859294",
    ),
    ResultSpec(
        key="retrieval_residual_scale050_type_success025",
        label="Train-state residual retrieval, scale 0.50, train family success >= 0.25",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_scale0p50_type_success0p25_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="train-split residual family reliability prior",
        pending_job="14859295/14859296",
    ),
    ResultSpec(
        key="retrieval_residual_scale050_type_success050",
        label="Train-state residual retrieval, scale 0.50, train family success >= 0.50",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_scale0p50_type_success0p50_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="train-split residual family reliability prior",
        pending_job="14859398/14859399",
    ),
    ResultSpec(
        key="retrieval_residual_scale050_type_success075",
        label="Train-state residual retrieval, scale 0.50, train family success >= 0.75",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_scale0p50_type_success0p75_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="train-split residual family reliability prior",
        pending_job="14859400/14859401",
    ),
    ResultSpec(
        key="retrieval_residual_scale025_type_success025",
        label="Train-state residual retrieval, scale 0.25, train family success >= 0.25",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_scale0p25_type_success0p25_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="train-split residual family reliability prior",
        pending_job="14859297/14859298",
    ),
    ResultSpec(
        key="retrieval_residual_scale035_type_success010",
        label="Train-state residual retrieval, scale 0.35, train family success >= 0.10",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_scale0p35_type_success010_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="repaired train-split reliability-prior diagnostic",
        pending_job="14862609/14862610",
    ),
    ResultSpec(
        key="retrieval_residual_scale035_type_success025",
        label="Train-state residual retrieval, scale 0.35, train family success >= 0.25",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_scale0p35_type_success025_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="repaired train-split reliability-prior diagnostic",
        pending_job="14862611/14862612",
    ),
    ResultSpec(
        key="retrieval_residual_scale075",
        label="Train-state residual retrieval, scale 0.75",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_scale0p75_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="tangent transport scale ablation",
        pending_job="14858879/14858880",
    ),
    ResultSpec(
        key="retrieval_residual_scale125",
        label="Train-state residual retrieval, scale 1.25",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_scale1p25_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="tangent transport scale ablation",
        pending_job="14858881/14858882",
    ),
    ResultSpec(
        key="retrieval_residual_hybrid_k32",
        label="Train-state residual + Gaussian proposals, K32 sigma0.35",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_hybrid_k32_sigma0p35_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="hybrid tangent/local proposal bridge",
        pending_job="14859042/14859043",
    ),
    ResultSpec(
        key="retrieval_residual_hybrid_k64",
        label="Train-state residual + Gaussian proposals, K64 sigma0.50",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_hybrid_k64_sigma0p50_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="hybrid tangent/local proposal bridge",
        pending_job="14859044/14859045",
    ),
    ResultSpec(
        key="retrieval_residual_knn4",
        label="KNN counterfactual residual retrieval",
        path="h16_policy_ckpt_near_miss_policy_bc5_bestpt_retrieval_residual_knn4_v2_summary.json",
        clean_deployment="yes",
        same_state_proposals="no",
        expert_proposal="no",
        story_role="KNN tangent proposal ablation",
        pending_job="14857114/14857115/14857116",
    ),
    ResultSpec(
        key="near_miss_only_lattice",
        label="Same-state lattice, near-miss only",
        path="h16_lattice_near_miss_only_v2_summary.json",
        clean_deployment="no",
        same_state_proposals="yes",
        expert_proposal="no",
        story_role="minimal mechanism result",
        fallback_success=0.5594,
    ),
    ResultSpec(
        key="no_expert_lattice",
        label="Same-state lattice, no expert",
        path="h16_lattice_no_expert_summary.json",
        clean_deployment="no",
        same_state_proposals="yes",
        expert_proposal="no",
        story_role="main conservative mechanism result",
        fallback_success=0.5699,
    ),
    ResultSpec(
        key="no_expert_lattice_policy_baseline_margin000",
        label="Same-state no-expert lattice with policy baseline candidate, margin 0.00",
        path="h16_lattice_no_expert_policy_baseline_margin000_summary.json",
        clean_deployment="no",
        same_state_proposals="yes",
        expert_proposal="no",
        story_role="negative policy-baseline abstention diagnostic",
        pending_job="14868661/14868662",
    ),
    ResultSpec(
        key="no_near_miss_no_expert_lattice",
        label="Same-state lattice, no expert/no near-miss",
        path="h16_lattice_no_near_miss_no_expert_v2_summary.json",
        clean_deployment="no",
        same_state_proposals="yes",
        expert_proposal="no",
        story_role="mechanism knockout",
        fallback_success=0.2557,
    ),
    ResultSpec(
        key="full_lattice",
        label="Same-state lattice, full",
        path="h16_lattice_summary.json",
        clean_deployment="no",
        same_state_proposals="yes",
        expert_proposal="yes",
        story_role="upper result with expert proposal",
        fallback_success=0.6933,
    ),
]


def main() -> int:
    RESULTS_DIR.mkdir(exist_ok=True)
    rows = [_row_for_spec(spec) for spec in SPECS]
    payload = {
        "baseline_h16_policy_success": BASELINE_H16_POLICY,
        "rows": rows,
        "best_clean": _best_row(rows, clean="yes"),
        "best_mechanism_no_expert": _best_mechanism_row(rows),
        "decision_notes": _decision_notes(rows),
    }
    json_path = RESULTS_DIR / "paper_table_status.json"
    md_path = RESULTS_DIR / "paper_table_status.md"
    json_path.write_text(json.dumps(payload, indent=2) + "\n")
    md_path.write_text(_render_markdown(payload) + "\n")
    print(f"Wrote {json_path}")
    print(f"Wrote {md_path}")
    return 0


def _row_for_spec(spec: ResultSpec) -> dict[str, Any]:
    path = RESULTS_DIR / spec.path if spec.path else None
    extracted = _extract_result(path) if path is not None and path.exists() else {}
    success = extracted.get("success", spec.fallback_success)
    if spec.clean_deployment == "diagnostic":
        success = extracted.get("candidate_oracle_success", success)
    path_exists = bool(path is not None and path.exists())
    status = "complete" if path_exists and success is not None else (
        "fallback" if success is not None else "pending"
    )
    return {
        **asdict(spec),
        "path_exists": path_exists,
        "status": status,
        "success": success,
        "std_success": extracted.get("std_success"),
        "completed_seeds": extracted.get("completed_seeds"),
        "num_completed": extracted.get("num_completed"),
        "best_config": extracted.get("best_config"),
        "gain_vs_h16_policy": (success - BASELINE_H16_POLICY) if success is not None else None,
    }


def _extract_result(path: Path) -> dict[str, Any]:
    data = json.loads(path.read_text())
    output: dict[str, Any] = {}
    if "mean_success" in data:
        output["std_success"] = _float_or_none(data.get("std_success"))
        output["num_completed"] = data.get("num_completed")
        output["candidate_oracle_success"] = _float_or_none(
            data.get("mean_candidate_oracle_success_rate")
        )
        if int(data.get("num_completed") or 0) > 0:
            output["success"] = _float_or_none(data.get("mean_success"))
    elif isinstance(data.get("best"), dict):
        best = data["best"]
        output["std_success"] = _float_or_none(best.get("std_success"))
        output["completed_seeds"] = best.get("completed_seeds")
        output["num_completed"] = best.get("num_completed")
        output["best_config"] = best.get("config")
        if int(best.get("num_completed") or 0) > 0:
            output["success"] = _float_or_none(best.get("mean_success"))
    elif "policy_rollout_success_rate" in data:
        output["num_completed"] = data.get("num_groups")
        if int(data.get("num_groups") or 0) > 0:
            output["success"] = _float_or_none(data.get("policy_rollout_success_rate"))
    return output


def _float_or_none(value: Any) -> float | None:
    if value is None:
        return None
    try:
        return float(value)
    except (TypeError, ValueError):
        return None


def _best_row(rows: list[dict[str, Any]], *, clean: str) -> dict[str, Any] | None:
    completed = [
        row
        for row in rows
        if row["clean_deployment"] == clean and row["success"] is not None
    ]
    return max(completed, key=lambda row: row["success"]) if completed else None


def _best_mechanism_row(rows: list[dict[str, Any]]) -> dict[str, Any] | None:
    candidates = [
        row
        for row in rows
        if row["same_state_proposals"] == "yes"
        and row["expert_proposal"] == "no"
        and row["success"] is not None
    ]
    return max(candidates, key=lambda row: row["success"]) if candidates else None


def _decision_notes(rows: list[dict[str, Any]]) -> list[str]:
    by_key = {row["key"]: row for row in rows}
    notes = [
        "Use no-expert same-state lattice as the conservative mechanism result, not as deployment-clean inference.",
        "Use full lattice only as an upper result because it includes expert proposals.",
        "Do not claim external SOTA from this table alone; add current external baselines separately.",
    ]
    clean_best = _best_row(rows, clean="yes")
    if clean_best is not None:
        notes.append(
            "Current best clean deployment row is "
            f"{clean_best['label']} at {_fmt_percent(clean_best['success'])}."
        )
    for key in (
        "field_optim",
        "retrieval_residual",
        "retrieval_residual_knn4",
        "retrieval_residual_k4_scale040_safe_margin020_mean_by_type",
    ):
        row = by_key[key]
        if row["success"] is None:
            notes.append(f"{row['label']} is pending ({row['pending_job']}).")
        elif row["success"] >= 0.40:
            notes.append(f"{row['label']} is strong enough to promote as a clean bridge.")
        elif row["success"] > 0.3293:
            notes.append(f"{row['label']} improves the clean bridge but is not yet the main result.")
        elif row["success"] > BASELINE_H16_POLICY:
            notes.append(
                f"{row['label']} is a positive clean bridge but remains below the current clean best."
            )
        else:
            notes.append(f"{row['label']} should be framed as a negative/diagnostic ablation.")
    policy_baseline = by_key["no_expert_lattice_policy_baseline_margin000"]
    no_expert = by_key["no_expert_lattice"]
    if policy_baseline["success"] is not None and no_expert["success"] is not None:
        notes.append(
            "Policy-baseline abstention inside the same-state no-expert lattice drops from "
            f"{_fmt_percent(no_expert['success'])} to {_fmt_percent(policy_baseline['success'])}, "
            "so the mechanism result should emphasize counterfactual proposal geometry rather than policy fallback."
        )
    nm_ties = [
        by_key["retrieval_residual_k4_composemasked_dropnmnoop_grid035040045_noopbonus003_nmbonus001"],
        by_key["retrieval_residual_k4_composemasked_dropnmnoop_grid035040045_noopbonus003_nmbonus002"],
    ]
    if clean_best is not None and all(row["success"] is not None for row in nm_ties):
        if all(abs(row["success"] - clean_best["success"]) < 1.0e-12 for row in nm_ties):
            notes.append(
                "Singleton near-miss priors 0.01/0.02 tie the clean best rather than improving it; "
                "the compatible chart preserves high-precision near-miss tangents, but they remain sparse."
            )
    oracle_prefix = by_key[
        "retrieval_residual_k4_composemasked_dropnmnoop_grid035040045_noopbonus003_oraclek8"
    ]
    if oracle_prefix["success"] is not None:
        notes.append(
            "Candidate-oracle prefix K=8 is diagnostic-only; compare it to the clean selected row "
            "to decide whether the next paper move should be proposal generation or field ranking."
        )
    oracle_trace = by_key[
        "retrieval_residual_k4_composemasked_dropnmnoop_grid035040045_noopbonus003_oraclek8trace"
    ]
    if oracle_trace["success"] is not None:
        notes.append(
            "Candidate-oracle branch trace is diagnostic-only; use its rank histogram and branch "
            "success profile to design one clean selector-calibration run rather than a broad sweep."
        )
    return notes


def _render_markdown(payload: dict[str, Any]) -> str:
    lines = [
        "# Paper Table Status",
        "",
        f"Baseline h=16 policy: {_fmt_percent(payload['baseline_h16_policy_success'])}",
        "",
        "| key | method | status | success | gain vs h16 | clean | same-state props | expert prop | role |",
        "|---|---|---|---:|---:|---|---|---|---|",
    ]
    for row in payload["rows"]:
        lines.append(
            "| {key} | {label} | {status} | {success} | {gain} | {clean} | {same} | {expert} | {role} |".format(
                key=row["key"],
                label=row["label"],
                status=_status_text(row),
                success=_fmt_percent(row["success"]),
                gain=_fmt_signed_percent(row["gain_vs_h16_policy"]),
                clean=row["clean_deployment"],
                same=row["same_state_proposals"],
                expert=row["expert_proposal"],
                role=row["story_role"],
            )
        )
    lines.extend(["", "## Decision Notes", ""])
    for note in payload["decision_notes"]:
        lines.append(f"- {note}")
    return "\n".join(lines)


def _status_text(row: dict[str, Any]) -> str:
    if row["status"] == "pending":
        return f"pending {row['pending_job']}".strip()
    if row["status"] == "fallback":
        return "fallback canonical"
    if row.get("best_config"):
        return f"complete {row['best_config']}"
    return "complete"


def _fmt_percent(value: float | None) -> str:
    return "pending" if value is None else f"{100.0 * value:.2f}%"


def _fmt_signed_percent(value: float | None) -> str:
    return "pending" if value is None else f"{100.0 * value:+.2f} pp"


if __name__ == "__main__":
    raise SystemExit(main())