{
  "run": "smoke_recurrent_vm_repair",
  "args": {
    "run_name": "smoke_recurrent_vm_repair",
    "model_name": "Qwen/Qwen3-4B",
    "seed": 173,
    "device": "",
    "seed_train_size": 16,
    "unlabeled_train_size": 32,
    "full_supervised_size": 32,
    "val_size": 16,
    "fresh_size": 16,
    "hard_size": 16,
    "max_arith_steps": 4,
    "qwen_batch_size": 8,
    "d_model": 128,
    "policy_d_model": 128,
    "compiler_layers": 1,
    "policy_layers": 1,
    "heads": 4,
    "dropout": 0.1,
    "batch_size": 16,
    "policy_batch_size": 32,
    "lr": 0.0005,
    "policy_lr": 0.0004,
    "dagger_lr": 0.0002,
    "answer_weight": 0.2,
    "seed_epochs": 1,
    "policy_epochs": 1,
    "dagger_epochs": 1,
    "full_supervised_epochs": 1,
    "teacher_max_steps": 8,
    "dagger_rollout_steps": 4,
    "search_topk": 2,
    "max_two_arg_pairs": 4,
    "k_values": [
      0,
      1,
      2,
      4,
      8
    ]
  },
  "results": [
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "teacher_policy",
      "split": "val_mixed",
      "mode": "base",
      "k": 0,
      "n": 16,
      "accuracy": 0.0,
      "valid_rate": 0.375,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 0.0,
      "base_accuracy": 0.0,
      "base_search_accuracy": 0.0,
      "base_oracle_accuracy": 0.0
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "teacher_policy",
      "split": "val_mixed",
      "mode": "oracle_teacher",
      "k": 0,
      "n": 16,
      "accuracy": 0.0,
      "valid_rate": 0.375,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 0.0,
      "base_accuracy": 0.0,
      "base_search_accuracy": 0.0,
      "base_oracle_accuracy": 0.0
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "teacher_policy",
      "split": "val_mixed",
      "mode": "oracle_teacher",
      "k": 1,
      "n": 16,
      "accuracy": 0.25,
      "valid_rate": 0.3125,
      "program_exact": 0.25,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 1.0,
      "base_accuracy": 0.0,
      "base_search_accuracy": 0.0,
      "base_oracle_accuracy": 0.0
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "teacher_policy",
      "split": "val_mixed",
      "mode": "oracle_teacher",
      "k": 2,
      "n": 16,
      "accuracy": 0.25,
      "valid_rate": 0.25,
      "program_exact": 0.25,
      "stop_rate": 0.25,
      "false_stop_rate": 0.0,
      "mean_steps": 1.75,
      "base_accuracy": 0.0,
      "base_search_accuracy": 0.0,
      "base_oracle_accuracy": 0.0
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "teacher_policy",
      "split": "val_mixed",
      "mode": "oracle_teacher",
      "k": 4,
      "n": 16,
      "accuracy": 0.25,
      "valid_rate": 0.25,
      "program_exact": 0.25,
      "stop_rate": 0.25,
      "false_stop_rate": 0.0,
      "mean_steps": 3.25,
      "base_accuracy": 0.0,
      "base_search_accuracy": 0.0,
      "base_oracle_accuracy": 0.0
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "teacher_policy",
      "split": "val_mixed",
      "mode": "oracle_teacher",
      "k": 8,
      "n": 16,
      "accuracy": 0.375,
      "valid_rate": 0.375,
      "program_exact": 0.25,
      "stop_rate": 0.375,
      "false_stop_rate": 0.0,
      "mean_steps": 6.0,
      "base_accuracy": 0.0,
      "base_search_accuracy": 0.0,
      "base_oracle_accuracy": 0.0
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "teacher_policy",
      "split": "val_mixed",
      "mode": "learned_stop",
      "k": 0,
      "n": 16,
      "accuracy": 0.0,
      "valid_rate": 0.375,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 0.0,
      "base_accuracy": 0.0,
      "base_search_accuracy": 0.0,
      "base_oracle_accuracy": 0.0
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "teacher_policy",
      "split": "val_mixed",
      "mode": "learned_stop",
      "k": 1,
      "n": 16,
      "accuracy": 0.0,
      "valid_rate": 0.375,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 1.0,
      "base_accuracy": 0.0,
      "base_search_accuracy": 0.0,
      "base_oracle_accuracy": 0.0
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "teacher_policy",
      "split": "val_mixed",
      "mode": "learned_stop",
      "k": 2,
      "n": 16,
      "accuracy": 0.0,
      "valid_rate": 0.375,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 2.0,
      "base_accuracy": 0.0,
      "base_search_accuracy": 0.0,
      "base_oracle_accuracy": 0.0
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "teacher_policy",
      "split": "val_mixed",
      "mode": "learned_stop",
      "k": 4,
      "n": 16,
      "accuracy": 0.0,
      "valid_rate": 0.375,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 4.0,
      "base_accuracy": 0.0,
      "base_search_accuracy": 0.0,
      "base_oracle_accuracy": 0.0
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "teacher_policy",
      "split": "val_mixed",
      "mode": "learned_stop",
      "k": 8,
      "n": 16,
      "accuracy": 0.0,
      "valid_rate": 0.375,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 8.0,
      "base_accuracy": 0.0,
      "base_search_accuracy": 0.0,
      "base_oracle_accuracy": 0.0
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "teacher_policy",
      "split": "val_mixed",
      "mode": "learned_forced",
      "k": 0,
      "n": 16,
      "accuracy": 0.0,
      "valid_rate": 0.375,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 0.0,
      "base_accuracy": 0.0,
      "base_search_accuracy": 0.0,
      "base_oracle_accuracy": 0.0
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "teacher_policy",
      "split": "val_mixed",
      "mode": "learned_forced",
      "k": 1,
      "n": 16,
      "accuracy": 0.0,
      "valid_rate": 0.375,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 1.0,
      "base_accuracy": 0.0,
      "base_search_accuracy": 0.0,
      "base_oracle_accuracy": 0.0
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "teacher_policy",
      "split": "val_mixed",
      "mode": "learned_forced",
      "k": 2,
      "n": 16,
      "accuracy": 0.0,
      "valid_rate": 0.375,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 2.0,
      "base_accuracy": 0.0,
      "base_search_accuracy": 0.0,
      "base_oracle_accuracy": 0.0
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "teacher_policy",
      "split": "val_mixed",
      "mode": "learned_forced",
      "k": 4,
      "n": 16,
      "accuracy": 0.0,
      "valid_rate": 0.375,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 4.0,
      "base_accuracy": 0.0,
      "base_search_accuracy": 0.0,
      "base_oracle_accuracy": 0.0
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "teacher_policy",
      "split": "val_mixed",
      "mode": "learned_forced",
      "k": 8,
      "n": 16,
      "accuracy": 0.0,
      "valid_rate": 0.375,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 8.0,
      "base_accuracy": 0.0,
      "base_search_accuracy": 0.0,
      "base_oracle_accuracy": 0.0
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "teacher_policy",
      "split": "fresh_standard",
      "mode": "base",
      "k": 0,
      "n": 16,
      "accuracy": 0.0,
      "valid_rate": 0.5625,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 0.0,
      "base_accuracy": 0.0,
      "base_search_accuracy": 0.0,
      "base_oracle_accuracy": 0.0
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "teacher_policy",
      "split": "fresh_standard",
      "mode": "oracle_teacher",
      "k": 0,
      "n": 16,
      "accuracy": 0.0,
      "valid_rate": 0.5625,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 0.0,
      "base_accuracy": 0.0,
      "base_search_accuracy": 0.0,
      "base_oracle_accuracy": 0.0
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "teacher_policy",
      "split": "fresh_standard",
      "mode": "oracle_teacher",
      "k": 1,
      "n": 16,
      "accuracy": 0.0625,
      "valid_rate": 0.5,
      "program_exact": 0.0625,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 1.0,
      "base_accuracy": 0.0,
      "base_search_accuracy": 0.0,
      "base_oracle_accuracy": 0.0
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "teacher_policy",
      "split": "fresh_standard",
      "mode": "oracle_teacher",
      "k": 2,
      "n": 16,
      "accuracy": 0.125,
      "valid_rate": 0.1875,
      "program_exact": 0.125,
      "stop_rate": 0.0625,
      "false_stop_rate": 0.0,
      "mean_steps": 1.9375,
      "base_accuracy": 0.0,
      "base_search_accuracy": 0.0,
      "base_oracle_accuracy": 0.0
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "teacher_policy",
      "split": "fresh_standard",
      "mode": "oracle_teacher",
      "k": 4,
      "n": 16,
      "accuracy": 0.125,
      "valid_rate": 0.1875,
      "program_exact": 0.125,
      "stop_rate": 0.125,
      "false_stop_rate": 0.0,
      "mean_steps": 3.6875,
      "base_accuracy": 0.0,
      "base_search_accuracy": 0.0,
      "base_oracle_accuracy": 0.0
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "teacher_policy",
      "split": "fresh_standard",
      "mode": "oracle_teacher",
      "k": 8,
      "n": 16,
      "accuracy": 0.3125,
      "valid_rate": 0.375,
      "program_exact": 0.125,
      "stop_rate": 0.3125,
      "false_stop_rate": 0.0,
      "mean_steps": 6.8125,
      "base_accuracy": 0.0,
      "base_search_accuracy": 0.0,
      "base_oracle_accuracy": 0.0
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "teacher_policy",
      "split": "fresh_standard",
      "mode": "learned_stop",
      "k": 0,
      "n": 16,
      "accuracy": 0.0,
      "valid_rate": 0.5625,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 0.0,
      "base_accuracy": 0.0,
      "base_search_accuracy": 0.0,
      "base_oracle_accuracy": 0.0
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "teacher_policy",
      "split": "fresh_standard",
      "mode": "learned_stop",
      "k": 1,
      "n": 16,
      "accuracy": 0.0,
      "valid_rate": 0.4375,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 1.0,
      "base_accuracy": 0.0,
      "base_search_accuracy": 0.0,
      "base_oracle_accuracy": 0.0
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "teacher_policy",
      "split": "fresh_standard",
      "mode": "learned_stop",
      "k": 2,
      "n": 16,
      "accuracy": 0.0,
      "valid_rate": 0.4375,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 2.0,
      "base_accuracy": 0.0,
      "base_search_accuracy": 0.0,
      "base_oracle_accuracy": 0.0
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "teacher_policy",
      "split": "fresh_standard",
      "mode": "learned_stop",
      "k": 4,
      "n": 16,
      "accuracy": 0.0,
      "valid_rate": 0.4375,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 4.0,
      "base_accuracy": 0.0,
      "base_search_accuracy": 0.0,
      "base_oracle_accuracy": 0.0
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "teacher_policy",
      "split": "fresh_standard",
      "mode": "learned_stop",
      "k": 8,
      "n": 16,
      "accuracy": 0.0,
      "valid_rate": 0.375,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 8.0,
      "base_accuracy": 0.0,
      "base_search_accuracy": 0.0,
      "base_oracle_accuracy": 0.0
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "teacher_policy",
      "split": "fresh_standard",
      "mode": "learned_forced",
      "k": 0,
      "n": 16,
      "accuracy": 0.0,
      "valid_rate": 0.5625,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 0.0,
      "base_accuracy": 0.0,
      "base_search_accuracy": 0.0,
      "base_oracle_accuracy": 0.0
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "teacher_policy",
      "split": "fresh_standard",
      "mode": "learned_forced",
      "k": 1,
      "n": 16,
      "accuracy": 0.0,
      "valid_rate": 0.4375,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 1.0,
      "base_accuracy": 0.0,
      "base_search_accuracy": 0.0,
      "base_oracle_accuracy": 0.0
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "teacher_policy",
      "split": "fresh_standard",
      "mode": "learned_forced",
      "k": 2,
      "n": 16,
      "accuracy": 0.0,
      "valid_rate": 0.4375,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 2.0,
      "base_accuracy": 0.0,
      "base_search_accuracy": 0.0,
      "base_oracle_accuracy": 0.0
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "teacher_policy",
      "split": "fresh_standard",
      "mode": "learned_forced",
      "k": 4,
      "n": 16,
      "accuracy": 0.0,
      "valid_rate": 0.4375,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 4.0,
      "base_accuracy": 0.0,
      "base_search_accuracy": 0.0,
      "base_oracle_accuracy": 0.0
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "teacher_policy",
      "split": "fresh_standard",
      "mode": "learned_forced",
      "k": 8,
      "n": 16,
      "accuracy": 0.0,
      "valid_rate": 0.375,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 8.0,
      "base_accuracy": 0.0,
      "base_search_accuracy": 0.0,
      "base_oracle_accuracy": 0.0
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "teacher_policy",
      "split": "fresh_paraphrase",
      "mode": "base",
      "k": 0,
      "n": 16,
      "accuracy": 0.0625,
      "valid_rate": 0.25,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 0.0,
      "base_accuracy": 0.0625,
      "base_search_accuracy": 0.0625,
      "base_oracle_accuracy": 0.0625
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "teacher_policy",
      "split": "fresh_paraphrase",
      "mode": "oracle_teacher",
      "k": 0,
      "n": 16,
      "accuracy": 0.0625,
      "valid_rate": 0.25,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 0.0,
      "base_accuracy": 0.0625,
      "base_search_accuracy": 0.0625,
      "base_oracle_accuracy": 0.0625
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "teacher_policy",
      "split": "fresh_paraphrase",
      "mode": "oracle_teacher",
      "k": 1,
      "n": 16,
      "accuracy": 0.125,
      "valid_rate": 0.25,
      "program_exact": 0.0625,
      "stop_rate": 0.0625,
      "false_stop_rate": 0.0,
      "mean_steps": 0.9375,
      "base_accuracy": 0.0625,
      "base_search_accuracy": 0.0625,
      "base_oracle_accuracy": 0.0625
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "teacher_policy",
      "split": "fresh_paraphrase",
      "mode": "oracle_teacher",
      "k": 2,
      "n": 16,
      "accuracy": 0.125,
      "valid_rate": 0.1875,
      "program_exact": 0.0625,
      "stop_rate": 0.125,
      "false_stop_rate": 0.0,
      "mean_steps": 1.8125,
      "base_accuracy": 0.0625,
      "base_search_accuracy": 0.0625,
      "base_oracle_accuracy": 0.0625
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "teacher_policy",
      "split": "fresh_paraphrase",
      "mode": "oracle_teacher",
      "k": 4,
      "n": 16,
      "accuracy": 0.125,
      "valid_rate": 0.125,
      "program_exact": 0.0625,
      "stop_rate": 0.125,
      "false_stop_rate": 0.0,
      "mean_steps": 3.5625,
      "base_accuracy": 0.0625,
      "base_search_accuracy": 0.0625,
      "base_oracle_accuracy": 0.0625
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "teacher_policy",
      "split": "fresh_paraphrase",
      "mode": "oracle_teacher",
      "k": 8,
      "n": 16,
      "accuracy": 0.1875,
      "valid_rate": 0.1875,
      "program_exact": 0.125,
      "stop_rate": 0.125,
      "false_stop_rate": 0.0,
      "mean_steps": 7.0625,
      "base_accuracy": 0.0625,
      "base_search_accuracy": 0.0625,
      "base_oracle_accuracy": 0.0625
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "teacher_policy",
      "split": "fresh_paraphrase",
      "mode": "learned_stop",
      "k": 0,
      "n": 16,
      "accuracy": 0.0625,
      "valid_rate": 0.25,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 0.0,
      "base_accuracy": 0.0625,
      "base_search_accuracy": 0.0625,
      "base_oracle_accuracy": 0.0625
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "teacher_policy",
      "split": "fresh_paraphrase",
      "mode": "learned_stop",
      "k": 1,
      "n": 16,
      "accuracy": 0.0625,
      "valid_rate": 0.25,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 1.0,
      "base_accuracy": 0.0625,
      "base_search_accuracy": 0.0625,
      "base_oracle_accuracy": 0.0625
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "teacher_policy",
      "split": "fresh_paraphrase",
      "mode": "learned_stop",
      "k": 2,
      "n": 16,
      "accuracy": 0.0625,
      "valid_rate": 0.25,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 2.0,
      "base_accuracy": 0.0625,
      "base_search_accuracy": 0.0625,
      "base_oracle_accuracy": 0.0625
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "teacher_policy",
      "split": "fresh_paraphrase",
      "mode": "learned_stop",
      "k": 4,
      "n": 16,
      "accuracy": 0.0625,
      "valid_rate": 0.25,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 4.0,
      "base_accuracy": 0.0625,
      "base_search_accuracy": 0.0625,
      "base_oracle_accuracy": 0.0625
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "teacher_policy",
      "split": "fresh_paraphrase",
      "mode": "learned_stop",
      "k": 8,
      "n": 16,
      "accuracy": 0.0,
      "valid_rate": 0.125,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 8.0,
      "base_accuracy": 0.0625,
      "base_search_accuracy": 0.0625,
      "base_oracle_accuracy": 0.0625
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "teacher_policy",
      "split": "fresh_paraphrase",
      "mode": "learned_forced",
      "k": 0,
      "n": 16,
      "accuracy": 0.0625,
      "valid_rate": 0.25,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 0.0,
      "base_accuracy": 0.0625,
      "base_search_accuracy": 0.0625,
      "base_oracle_accuracy": 0.0625
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "teacher_policy",
      "split": "fresh_paraphrase",
      "mode": "learned_forced",
      "k": 1,
      "n": 16,
      "accuracy": 0.0625,
      "valid_rate": 0.25,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 1.0,
      "base_accuracy": 0.0625,
      "base_search_accuracy": 0.0625,
      "base_oracle_accuracy": 0.0625
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "teacher_policy",
      "split": "fresh_paraphrase",
      "mode": "learned_forced",
      "k": 2,
      "n": 16,
      "accuracy": 0.0625,
      "valid_rate": 0.25,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 2.0,
      "base_accuracy": 0.0625,
      "base_search_accuracy": 0.0625,
      "base_oracle_accuracy": 0.0625
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "teacher_policy",
      "split": "fresh_paraphrase",
      "mode": "learned_forced",
      "k": 4,
      "n": 16,
      "accuracy": 0.0625,
      "valid_rate": 0.25,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 4.0,
      "base_accuracy": 0.0625,
      "base_search_accuracy": 0.0625,
      "base_oracle_accuracy": 0.0625
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "teacher_policy",
      "split": "fresh_paraphrase",
      "mode": "learned_forced",
      "k": 8,
      "n": 16,
      "accuracy": 0.0,
      "valid_rate": 0.125,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 8.0,
      "base_accuracy": 0.0625,
      "base_search_accuracy": 0.0625,
      "base_oracle_accuracy": 0.0625
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "teacher_policy",
      "split": "fresh_paired",
      "mode": "base",
      "k": 0,
      "n": 16,
      "accuracy": 0.0625,
      "valid_rate": 0.6875,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 0.0,
      "base_accuracy": 0.0625,
      "base_search_accuracy": 0.0625,
      "base_oracle_accuracy": 0.0625
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "teacher_policy",
      "split": "fresh_paired",
      "mode": "oracle_teacher",
      "k": 0,
      "n": 16,
      "accuracy": 0.0625,
      "valid_rate": 0.6875,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 0.0,
      "base_accuracy": 0.0625,
      "base_search_accuracy": 0.0625,
      "base_oracle_accuracy": 0.0625
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "teacher_policy",
      "split": "fresh_paired",
      "mode": "oracle_teacher",
      "k": 1,
      "n": 16,
      "accuracy": 0.1875,
      "valid_rate": 0.625,
      "program_exact": 0.125,
      "stop_rate": 0.0625,
      "false_stop_rate": 0.0,
      "mean_steps": 0.9375,
      "base_accuracy": 0.0625,
      "base_search_accuracy": 0.0625,
      "base_oracle_accuracy": 0.0625
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "teacher_policy",
      "split": "fresh_paired",
      "mode": "oracle_teacher",
      "k": 2,
      "n": 16,
      "accuracy": 0.4375,
      "valid_rate": 0.5625,
      "program_exact": 0.375,
      "stop_rate": 0.1875,
      "false_stop_rate": 0.0,
      "mean_steps": 1.75,
      "base_accuracy": 0.0625,
      "base_search_accuracy": 0.0625,
      "base_oracle_accuracy": 0.0625
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "teacher_policy",
      "split": "fresh_paired",
      "mode": "oracle_teacher",
      "k": 4,
      "n": 16,
      "accuracy": 0.4375,
      "valid_rate": 0.4375,
      "program_exact": 0.375,
      "stop_rate": 0.4375,
      "false_stop_rate": 0.0,
      "mean_steps": 2.875,
      "base_accuracy": 0.0625,
      "base_search_accuracy": 0.0625,
      "base_oracle_accuracy": 0.0625
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "teacher_policy",
      "split": "fresh_paired",
      "mode": "oracle_teacher",
      "k": 8,
      "n": 16,
      "accuracy": 0.5625,
      "valid_rate": 0.625,
      "program_exact": 0.375,
      "stop_rate": 0.5625,
      "false_stop_rate": 0.0,
      "mean_steps": 4.8125,
      "base_accuracy": 0.0625,
      "base_search_accuracy": 0.0625,
      "base_oracle_accuracy": 0.0625
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "teacher_policy",
      "split": "fresh_paired",
      "mode": "learned_stop",
      "k": 0,
      "n": 16,
      "accuracy": 0.0625,
      "valid_rate": 0.6875,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 0.0,
      "base_accuracy": 0.0625,
      "base_search_accuracy": 0.0625,
      "base_oracle_accuracy": 0.0625
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "teacher_policy",
      "split": "fresh_paired",
      "mode": "learned_stop",
      "k": 1,
      "n": 16,
      "accuracy": 0.0625,
      "valid_rate": 0.6875,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 1.0,
      "base_accuracy": 0.0625,
      "base_search_accuracy": 0.0625,
      "base_oracle_accuracy": 0.0625
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "teacher_policy",
      "split": "fresh_paired",
      "mode": "learned_stop",
      "k": 2,
      "n": 16,
      "accuracy": 0.0625,
      "valid_rate": 0.6875,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 2.0,
      "base_accuracy": 0.0625,
      "base_search_accuracy": 0.0625,
      "base_oracle_accuracy": 0.0625
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "teacher_policy",
      "split": "fresh_paired",
      "mode": "learned_stop",
      "k": 4,
      "n": 16,
      "accuracy": 0.0625,
      "valid_rate": 0.6875,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 4.0,
      "base_accuracy": 0.0625,
      "base_search_accuracy": 0.0625,
      "base_oracle_accuracy": 0.0625
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "teacher_policy",
      "split": "fresh_paired",
      "mode": "learned_stop",
      "k": 8,
      "n": 16,
      "accuracy": 0.0,
      "valid_rate": 0.5625,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 8.0,
      "base_accuracy": 0.0625,
      "base_search_accuracy": 0.0625,
      "base_oracle_accuracy": 0.0625
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "teacher_policy",
      "split": "fresh_paired",
      "mode": "learned_forced",
      "k": 0,
      "n": 16,
      "accuracy": 0.0625,
      "valid_rate": 0.6875,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 0.0,
      "base_accuracy": 0.0625,
      "base_search_accuracy": 0.0625,
      "base_oracle_accuracy": 0.0625
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "teacher_policy",
      "split": "fresh_paired",
      "mode": "learned_forced",
      "k": 1,
      "n": 16,
      "accuracy": 0.0625,
      "valid_rate": 0.6875,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 1.0,
      "base_accuracy": 0.0625,
      "base_search_accuracy": 0.0625,
      "base_oracle_accuracy": 0.0625
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "teacher_policy",
      "split": "fresh_paired",
      "mode": "learned_forced",
      "k": 2,
      "n": 16,
      "accuracy": 0.0625,
      "valid_rate": 0.6875,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 2.0,
      "base_accuracy": 0.0625,
      "base_search_accuracy": 0.0625,
      "base_oracle_accuracy": 0.0625
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "teacher_policy",
      "split": "fresh_paired",
      "mode": "learned_forced",
      "k": 4,
      "n": 16,
      "accuracy": 0.0625,
      "valid_rate": 0.6875,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 4.0,
      "base_accuracy": 0.0625,
      "base_search_accuracy": 0.0625,
      "base_oracle_accuracy": 0.0625
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "teacher_policy",
      "split": "fresh_paired",
      "mode": "learned_forced",
      "k": 8,
      "n": 16,
      "accuracy": 0.0,
      "valid_rate": 0.5625,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 8.0,
      "base_accuracy": 0.0625,
      "base_search_accuracy": 0.0625,
      "base_oracle_accuracy": 0.0625
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "teacher_policy",
      "split": "hard_composition",
      "mode": "base",
      "k": 0,
      "n": 16,
      "accuracy": 0.0625,
      "valid_rate": 0.5625,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 0.0,
      "base_accuracy": 0.0625,
      "base_search_accuracy": 0.0625,
      "base_oracle_accuracy": 0.0625
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "teacher_policy",
      "split": "hard_composition",
      "mode": "oracle_teacher",
      "k": 0,
      "n": 16,
      "accuracy": 0.0625,
      "valid_rate": 0.5625,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 0.0,
      "base_accuracy": 0.0625,
      "base_search_accuracy": 0.0625,
      "base_oracle_accuracy": 0.0625
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "teacher_policy",
      "split": "hard_composition",
      "mode": "oracle_teacher",
      "k": 1,
      "n": 16,
      "accuracy": 0.0625,
      "valid_rate": 0.5,
      "program_exact": 0.0,
      "stop_rate": 0.0625,
      "false_stop_rate": 0.0,
      "mean_steps": 0.9375,
      "base_accuracy": 0.0625,
      "base_search_accuracy": 0.0625,
      "base_oracle_accuracy": 0.0625
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "teacher_policy",
      "split": "hard_composition",
      "mode": "oracle_teacher",
      "k": 2,
      "n": 16,
      "accuracy": 0.1875,
      "valid_rate": 0.3125,
      "program_exact": 0.125,
      "stop_rate": 0.0625,
      "false_stop_rate": 0.0,
      "mean_steps": 1.875,
      "base_accuracy": 0.0625,
      "base_search_accuracy": 0.0625,
      "base_oracle_accuracy": 0.0625
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "teacher_policy",
      "split": "hard_composition",
      "mode": "oracle_teacher",
      "k": 4,
      "n": 16,
      "accuracy": 0.1875,
      "valid_rate": 0.25,
      "program_exact": 0.125,
      "stop_rate": 0.1875,
      "false_stop_rate": 0.0,
      "mean_steps": 3.5,
      "base_accuracy": 0.0625,
      "base_search_accuracy": 0.0625,
      "base_oracle_accuracy": 0.0625
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "teacher_policy",
      "split": "hard_composition",
      "mode": "oracle_teacher",
      "k": 8,
      "n": 16,
      "accuracy": 0.3125,
      "valid_rate": 0.5,
      "program_exact": 0.25,
      "stop_rate": 0.1875,
      "false_stop_rate": 0.0,
      "mean_steps": 6.75,
      "base_accuracy": 0.0625,
      "base_search_accuracy": 0.0625,
      "base_oracle_accuracy": 0.0625
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "teacher_policy",
      "split": "hard_composition",
      "mode": "learned_stop",
      "k": 0,
      "n": 16,
      "accuracy": 0.0625,
      "valid_rate": 0.5625,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 0.0,
      "base_accuracy": 0.0625,
      "base_search_accuracy": 0.0625,
      "base_oracle_accuracy": 0.0625
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "teacher_policy",
      "split": "hard_composition",
      "mode": "learned_stop",
      "k": 1,
      "n": 16,
      "accuracy": 0.0625,
      "valid_rate": 0.5,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 1.0,
      "base_accuracy": 0.0625,
      "base_search_accuracy": 0.0625,
      "base_oracle_accuracy": 0.0625
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "teacher_policy",
      "split": "hard_composition",
      "mode": "learned_stop",
      "k": 2,
      "n": 16,
      "accuracy": 0.0625,
      "valid_rate": 0.5,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 2.0,
      "base_accuracy": 0.0625,
      "base_search_accuracy": 0.0625,
      "base_oracle_accuracy": 0.0625
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "teacher_policy",
      "split": "hard_composition",
      "mode": "learned_stop",
      "k": 4,
      "n": 16,
      "accuracy": 0.0,
      "valid_rate": 0.4375,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 4.0,
      "base_accuracy": 0.0625,
      "base_search_accuracy": 0.0625,
      "base_oracle_accuracy": 0.0625
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "teacher_policy",
      "split": "hard_composition",
      "mode": "learned_stop",
      "k": 8,
      "n": 16,
      "accuracy": 0.0,
      "valid_rate": 0.4375,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 8.0,
      "base_accuracy": 0.0625,
      "base_search_accuracy": 0.0625,
      "base_oracle_accuracy": 0.0625
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "teacher_policy",
      "split": "hard_composition",
      "mode": "learned_forced",
      "k": 0,
      "n": 16,
      "accuracy": 0.0625,
      "valid_rate": 0.5625,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 0.0,
      "base_accuracy": 0.0625,
      "base_search_accuracy": 0.0625,
      "base_oracle_accuracy": 0.0625
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "teacher_policy",
      "split": "hard_composition",
      "mode": "learned_forced",
      "k": 1,
      "n": 16,
      "accuracy": 0.0625,
      "valid_rate": 0.5,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 1.0,
      "base_accuracy": 0.0625,
      "base_search_accuracy": 0.0625,
      "base_oracle_accuracy": 0.0625
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "teacher_policy",
      "split": "hard_composition",
      "mode": "learned_forced",
      "k": 2,
      "n": 16,
      "accuracy": 0.0625,
      "valid_rate": 0.5,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 2.0,
      "base_accuracy": 0.0625,
      "base_search_accuracy": 0.0625,
      "base_oracle_accuracy": 0.0625
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "teacher_policy",
      "split": "hard_composition",
      "mode": "learned_forced",
      "k": 4,
      "n": 16,
      "accuracy": 0.0625,
      "valid_rate": 0.5,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 4.0,
      "base_accuracy": 0.0625,
      "base_search_accuracy": 0.0625,
      "base_oracle_accuracy": 0.0625
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "teacher_policy",
      "split": "hard_composition",
      "mode": "learned_forced",
      "k": 8,
      "n": 16,
      "accuracy": 0.0,
      "valid_rate": 0.4375,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 8.0,
      "base_accuracy": 0.0625,
      "base_search_accuracy": 0.0625,
      "base_oracle_accuracy": 0.0625
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "dagger_policy",
      "split": "val_mixed",
      "mode": "base",
      "k": 0,
      "n": 16,
      "accuracy": 0.0,
      "valid_rate": 0.375,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 0.0,
      "base_accuracy": 0.0,
      "base_search_accuracy": 0.0,
      "base_oracle_accuracy": 0.0
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "dagger_policy",
      "split": "val_mixed",
      "mode": "oracle_teacher",
      "k": 0,
      "n": 16,
      "accuracy": 0.0,
      "valid_rate": 0.375,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 0.0,
      "base_accuracy": 0.0,
      "base_search_accuracy": 0.0,
      "base_oracle_accuracy": 0.0
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "dagger_policy",
      "split": "val_mixed",
      "mode": "oracle_teacher",
      "k": 1,
      "n": 16,
      "accuracy": 0.25,
      "valid_rate": 0.3125,
      "program_exact": 0.25,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 1.0,
      "base_accuracy": 0.0,
      "base_search_accuracy": 0.0,
      "base_oracle_accuracy": 0.0
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "dagger_policy",
      "split": "val_mixed",
      "mode": "oracle_teacher",
      "k": 2,
      "n": 16,
      "accuracy": 0.25,
      "valid_rate": 0.25,
      "program_exact": 0.25,
      "stop_rate": 0.25,
      "false_stop_rate": 0.0,
      "mean_steps": 1.75,
      "base_accuracy": 0.0,
      "base_search_accuracy": 0.0,
      "base_oracle_accuracy": 0.0
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "dagger_policy",
      "split": "val_mixed",
      "mode": "oracle_teacher",
      "k": 4,
      "n": 16,
      "accuracy": 0.25,
      "valid_rate": 0.25,
      "program_exact": 0.25,
      "stop_rate": 0.25,
      "false_stop_rate": 0.0,
      "mean_steps": 3.25,
      "base_accuracy": 0.0,
      "base_search_accuracy": 0.0,
      "base_oracle_accuracy": 0.0
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "dagger_policy",
      "split": "val_mixed",
      "mode": "oracle_teacher",
      "k": 8,
      "n": 16,
      "accuracy": 0.375,
      "valid_rate": 0.375,
      "program_exact": 0.25,
      "stop_rate": 0.375,
      "false_stop_rate": 0.0,
      "mean_steps": 6.0,
      "base_accuracy": 0.0,
      "base_search_accuracy": 0.0,
      "base_oracle_accuracy": 0.0
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "dagger_policy",
      "split": "val_mixed",
      "mode": "learned_stop",
      "k": 0,
      "n": 16,
      "accuracy": 0.0,
      "valid_rate": 0.375,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 0.0,
      "base_accuracy": 0.0,
      "base_search_accuracy": 0.0,
      "base_oracle_accuracy": 0.0
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "dagger_policy",
      "split": "val_mixed",
      "mode": "learned_stop",
      "k": 1,
      "n": 16,
      "accuracy": 0.0,
      "valid_rate": 0.125,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 1.0,
      "base_accuracy": 0.0,
      "base_search_accuracy": 0.0,
      "base_oracle_accuracy": 0.0
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "dagger_policy",
      "split": "val_mixed",
      "mode": "learned_stop",
      "k": 2,
      "n": 16,
      "accuracy": 0.0,
      "valid_rate": 0.0625,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 2.0,
      "base_accuracy": 0.0,
      "base_search_accuracy": 0.0,
      "base_oracle_accuracy": 0.0
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "dagger_policy",
      "split": "val_mixed",
      "mode": "learned_stop",
      "k": 4,
      "n": 16,
      "accuracy": 0.0,
      "valid_rate": 0.0625,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 4.0,
      "base_accuracy": 0.0,
      "base_search_accuracy": 0.0,
      "base_oracle_accuracy": 0.0
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "dagger_policy",
      "split": "val_mixed",
      "mode": "learned_stop",
      "k": 8,
      "n": 16,
      "accuracy": 0.0,
      "valid_rate": 0.0625,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 8.0,
      "base_accuracy": 0.0,
      "base_search_accuracy": 0.0,
      "base_oracle_accuracy": 0.0
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "dagger_policy",
      "split": "val_mixed",
      "mode": "learned_forced",
      "k": 0,
      "n": 16,
      "accuracy": 0.0,
      "valid_rate": 0.375,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 0.0,
      "base_accuracy": 0.0,
      "base_search_accuracy": 0.0,
      "base_oracle_accuracy": 0.0
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "dagger_policy",
      "split": "val_mixed",
      "mode": "learned_forced",
      "k": 1,
      "n": 16,
      "accuracy": 0.0,
      "valid_rate": 0.125,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 1.0,
      "base_accuracy": 0.0,
      "base_search_accuracy": 0.0,
      "base_oracle_accuracy": 0.0
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "dagger_policy",
      "split": "val_mixed",
      "mode": "learned_forced",
      "k": 2,
      "n": 16,
      "accuracy": 0.0,
      "valid_rate": 0.125,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 2.0,
      "base_accuracy": 0.0,
      "base_search_accuracy": 0.0,
      "base_oracle_accuracy": 0.0
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "dagger_policy",
      "split": "val_mixed",
      "mode": "learned_forced",
      "k": 4,
      "n": 16,
      "accuracy": 0.0,
      "valid_rate": 0.0625,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 4.0,
      "base_accuracy": 0.0,
      "base_search_accuracy": 0.0,
      "base_oracle_accuracy": 0.0
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "dagger_policy",
      "split": "val_mixed",
      "mode": "learned_forced",
      "k": 8,
      "n": 16,
      "accuracy": 0.0,
      "valid_rate": 0.0625,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 8.0,
      "base_accuracy": 0.0,
      "base_search_accuracy": 0.0,
      "base_oracle_accuracy": 0.0
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "dagger_policy",
      "split": "fresh_standard",
      "mode": "base",
      "k": 0,
      "n": 16,
      "accuracy": 0.0,
      "valid_rate": 0.5625,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 0.0,
      "base_accuracy": 0.0,
      "base_search_accuracy": 0.0,
      "base_oracle_accuracy": 0.0
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "dagger_policy",
      "split": "fresh_standard",
      "mode": "oracle_teacher",
      "k": 0,
      "n": 16,
      "accuracy": 0.0,
      "valid_rate": 0.5625,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 0.0,
      "base_accuracy": 0.0,
      "base_search_accuracy": 0.0,
      "base_oracle_accuracy": 0.0
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "dagger_policy",
      "split": "fresh_standard",
      "mode": "oracle_teacher",
      "k": 1,
      "n": 16,
      "accuracy": 0.0625,
      "valid_rate": 0.5,
      "program_exact": 0.0625,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 1.0,
      "base_accuracy": 0.0,
      "base_search_accuracy": 0.0,
      "base_oracle_accuracy": 0.0
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "dagger_policy",
      "split": "fresh_standard",
      "mode": "oracle_teacher",
      "k": 2,
      "n": 16,
      "accuracy": 0.125,
      "valid_rate": 0.1875,
      "program_exact": 0.125,
      "stop_rate": 0.0625,
      "false_stop_rate": 0.0,
      "mean_steps": 1.9375,
      "base_accuracy": 0.0,
      "base_search_accuracy": 0.0,
      "base_oracle_accuracy": 0.0
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "dagger_policy",
      "split": "fresh_standard",
      "mode": "oracle_teacher",
      "k": 4,
      "n": 16,
      "accuracy": 0.125,
      "valid_rate": 0.1875,
      "program_exact": 0.125,
      "stop_rate": 0.125,
      "false_stop_rate": 0.0,
      "mean_steps": 3.6875,
      "base_accuracy": 0.0,
      "base_search_accuracy": 0.0,
      "base_oracle_accuracy": 0.0
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "dagger_policy",
      "split": "fresh_standard",
      "mode": "oracle_teacher",
      "k": 8,
      "n": 16,
      "accuracy": 0.3125,
      "valid_rate": 0.375,
      "program_exact": 0.125,
      "stop_rate": 0.3125,
      "false_stop_rate": 0.0,
      "mean_steps": 6.8125,
      "base_accuracy": 0.0,
      "base_search_accuracy": 0.0,
      "base_oracle_accuracy": 0.0
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "dagger_policy",
      "split": "fresh_standard",
      "mode": "learned_stop",
      "k": 0,
      "n": 16,
      "accuracy": 0.0,
      "valid_rate": 0.5625,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 0.0,
      "base_accuracy": 0.0,
      "base_search_accuracy": 0.0,
      "base_oracle_accuracy": 0.0
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "dagger_policy",
      "split": "fresh_standard",
      "mode": "learned_stop",
      "k": 1,
      "n": 16,
      "accuracy": 0.0,
      "valid_rate": 0.375,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 1.0,
      "base_accuracy": 0.0,
      "base_search_accuracy": 0.0,
      "base_oracle_accuracy": 0.0
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "dagger_policy",
      "split": "fresh_standard",
      "mode": "learned_stop",
      "k": 2,
      "n": 16,
      "accuracy": 0.0,
      "valid_rate": 0.4375,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 2.0,
      "base_accuracy": 0.0,
      "base_search_accuracy": 0.0,
      "base_oracle_accuracy": 0.0
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "dagger_policy",
      "split": "fresh_standard",
      "mode": "learned_stop",
      "k": 4,
      "n": 16,
      "accuracy": 0.0,
      "valid_rate": 0.375,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 4.0,
      "base_accuracy": 0.0,
      "base_search_accuracy": 0.0,
      "base_oracle_accuracy": 0.0
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "dagger_policy",
      "split": "fresh_standard",
      "mode": "learned_stop",
      "k": 8,
      "n": 16,
      "accuracy": 0.0,
      "valid_rate": 0.375,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 8.0,
      "base_accuracy": 0.0,
      "base_search_accuracy": 0.0,
      "base_oracle_accuracy": 0.0
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "dagger_policy",
      "split": "fresh_standard",
      "mode": "learned_forced",
      "k": 0,
      "n": 16,
      "accuracy": 0.0,
      "valid_rate": 0.5625,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 0.0,
      "base_accuracy": 0.0,
      "base_search_accuracy": 0.0,
      "base_oracle_accuracy": 0.0
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "dagger_policy",
      "split": "fresh_standard",
      "mode": "learned_forced",
      "k": 1,
      "n": 16,
      "accuracy": 0.0,
      "valid_rate": 0.5,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 1.0,
      "base_accuracy": 0.0,
      "base_search_accuracy": 0.0,
      "base_oracle_accuracy": 0.0
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "dagger_policy",
      "split": "fresh_standard",
      "mode": "learned_forced",
      "k": 2,
      "n": 16,
      "accuracy": 0.0,
      "valid_rate": 0.375,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 2.0,
      "base_accuracy": 0.0,
      "base_search_accuracy": 0.0,
      "base_oracle_accuracy": 0.0
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "dagger_policy",
      "split": "fresh_standard",
      "mode": "learned_forced",
      "k": 4,
      "n": 16,
      "accuracy": 0.0,
      "valid_rate": 0.375,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 4.0,
      "base_accuracy": 0.0,
      "base_search_accuracy": 0.0,
      "base_oracle_accuracy": 0.0
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "dagger_policy",
      "split": "fresh_standard",
      "mode": "learned_forced",
      "k": 8,
      "n": 16,
      "accuracy": 0.0,
      "valid_rate": 0.375,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 8.0,
      "base_accuracy": 0.0,
      "base_search_accuracy": 0.0,
      "base_oracle_accuracy": 0.0
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "dagger_policy",
      "split": "fresh_paraphrase",
      "mode": "base",
      "k": 0,
      "n": 16,
      "accuracy": 0.0625,
      "valid_rate": 0.25,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 0.0,
      "base_accuracy": 0.0625,
      "base_search_accuracy": 0.0625,
      "base_oracle_accuracy": 0.0625
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "dagger_policy",
      "split": "fresh_paraphrase",
      "mode": "oracle_teacher",
      "k": 0,
      "n": 16,
      "accuracy": 0.0625,
      "valid_rate": 0.25,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 0.0,
      "base_accuracy": 0.0625,
      "base_search_accuracy": 0.0625,
      "base_oracle_accuracy": 0.0625
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "dagger_policy",
      "split": "fresh_paraphrase",
      "mode": "oracle_teacher",
      "k": 1,
      "n": 16,
      "accuracy": 0.125,
      "valid_rate": 0.25,
      "program_exact": 0.0625,
      "stop_rate": 0.0625,
      "false_stop_rate": 0.0,
      "mean_steps": 0.9375,
      "base_accuracy": 0.0625,
      "base_search_accuracy": 0.0625,
      "base_oracle_accuracy": 0.0625
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "dagger_policy",
      "split": "fresh_paraphrase",
      "mode": "oracle_teacher",
      "k": 2,
      "n": 16,
      "accuracy": 0.125,
      "valid_rate": 0.1875,
      "program_exact": 0.0625,
      "stop_rate": 0.125,
      "false_stop_rate": 0.0,
      "mean_steps": 1.8125,
      "base_accuracy": 0.0625,
      "base_search_accuracy": 0.0625,
      "base_oracle_accuracy": 0.0625
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "dagger_policy",
      "split": "fresh_paraphrase",
      "mode": "oracle_teacher",
      "k": 4,
      "n": 16,
      "accuracy": 0.125,
      "valid_rate": 0.125,
      "program_exact": 0.0625,
      "stop_rate": 0.125,
      "false_stop_rate": 0.0,
      "mean_steps": 3.5625,
      "base_accuracy": 0.0625,
      "base_search_accuracy": 0.0625,
      "base_oracle_accuracy": 0.0625
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "dagger_policy",
      "split": "fresh_paraphrase",
      "mode": "oracle_teacher",
      "k": 8,
      "n": 16,
      "accuracy": 0.1875,
      "valid_rate": 0.1875,
      "program_exact": 0.125,
      "stop_rate": 0.125,
      "false_stop_rate": 0.0,
      "mean_steps": 7.0625,
      "base_accuracy": 0.0625,
      "base_search_accuracy": 0.0625,
      "base_oracle_accuracy": 0.0625
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "dagger_policy",
      "split": "fresh_paraphrase",
      "mode": "learned_stop",
      "k": 0,
      "n": 16,
      "accuracy": 0.0625,
      "valid_rate": 0.25,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 0.0,
      "base_accuracy": 0.0625,
      "base_search_accuracy": 0.0625,
      "base_oracle_accuracy": 0.0625
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "dagger_policy",
      "split": "fresh_paraphrase",
      "mode": "learned_stop",
      "k": 1,
      "n": 16,
      "accuracy": 0.0625,
      "valid_rate": 0.1875,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 1.0,
      "base_accuracy": 0.0625,
      "base_search_accuracy": 0.0625,
      "base_oracle_accuracy": 0.0625
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "dagger_policy",
      "split": "fresh_paraphrase",
      "mode": "learned_stop",
      "k": 2,
      "n": 16,
      "accuracy": 0.0625,
      "valid_rate": 0.1875,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 2.0,
      "base_accuracy": 0.0625,
      "base_search_accuracy": 0.0625,
      "base_oracle_accuracy": 0.0625
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "dagger_policy",
      "split": "fresh_paraphrase",
      "mode": "learned_stop",
      "k": 4,
      "n": 16,
      "accuracy": 0.0625,
      "valid_rate": 0.1875,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 4.0,
      "base_accuracy": 0.0625,
      "base_search_accuracy": 0.0625,
      "base_oracle_accuracy": 0.0625
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "dagger_policy",
      "split": "fresh_paraphrase",
      "mode": "learned_stop",
      "k": 8,
      "n": 16,
      "accuracy": 0.0625,
      "valid_rate": 0.1875,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 8.0,
      "base_accuracy": 0.0625,
      "base_search_accuracy": 0.0625,
      "base_oracle_accuracy": 0.0625
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "dagger_policy",
      "split": "fresh_paraphrase",
      "mode": "learned_forced",
      "k": 0,
      "n": 16,
      "accuracy": 0.0625,
      "valid_rate": 0.25,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 0.0,
      "base_accuracy": 0.0625,
      "base_search_accuracy": 0.0625,
      "base_oracle_accuracy": 0.0625
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "dagger_policy",
      "split": "fresh_paraphrase",
      "mode": "learned_forced",
      "k": 1,
      "n": 16,
      "accuracy": 0.0625,
      "valid_rate": 0.1875,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 1.0,
      "base_accuracy": 0.0625,
      "base_search_accuracy": 0.0625,
      "base_oracle_accuracy": 0.0625
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "dagger_policy",
      "split": "fresh_paraphrase",
      "mode": "learned_forced",
      "k": 2,
      "n": 16,
      "accuracy": 0.0625,
      "valid_rate": 0.1875,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 2.0,
      "base_accuracy": 0.0625,
      "base_search_accuracy": 0.0625,
      "base_oracle_accuracy": 0.0625
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "dagger_policy",
      "split": "fresh_paraphrase",
      "mode": "learned_forced",
      "k": 4,
      "n": 16,
      "accuracy": 0.0625,
      "valid_rate": 0.1875,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 4.0,
      "base_accuracy": 0.0625,
      "base_search_accuracy": 0.0625,
      "base_oracle_accuracy": 0.0625
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "dagger_policy",
      "split": "fresh_paraphrase",
      "mode": "learned_forced",
      "k": 8,
      "n": 16,
      "accuracy": 0.0625,
      "valid_rate": 0.1875,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 8.0,
      "base_accuracy": 0.0625,
      "base_search_accuracy": 0.0625,
      "base_oracle_accuracy": 0.0625
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "dagger_policy",
      "split": "fresh_paired",
      "mode": "base",
      "k": 0,
      "n": 16,
      "accuracy": 0.0625,
      "valid_rate": 0.6875,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 0.0,
      "base_accuracy": 0.0625,
      "base_search_accuracy": 0.0625,
      "base_oracle_accuracy": 0.0625
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "dagger_policy",
      "split": "fresh_paired",
      "mode": "oracle_teacher",
      "k": 0,
      "n": 16,
      "accuracy": 0.0625,
      "valid_rate": 0.6875,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 0.0,
      "base_accuracy": 0.0625,
      "base_search_accuracy": 0.0625,
      "base_oracle_accuracy": 0.0625
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "dagger_policy",
      "split": "fresh_paired",
      "mode": "oracle_teacher",
      "k": 1,
      "n": 16,
      "accuracy": 0.1875,
      "valid_rate": 0.625,
      "program_exact": 0.125,
      "stop_rate": 0.0625,
      "false_stop_rate": 0.0,
      "mean_steps": 0.9375,
      "base_accuracy": 0.0625,
      "base_search_accuracy": 0.0625,
      "base_oracle_accuracy": 0.0625
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "dagger_policy",
      "split": "fresh_paired",
      "mode": "oracle_teacher",
      "k": 2,
      "n": 16,
      "accuracy": 0.4375,
      "valid_rate": 0.5625,
      "program_exact": 0.375,
      "stop_rate": 0.1875,
      "false_stop_rate": 0.0,
      "mean_steps": 1.75,
      "base_accuracy": 0.0625,
      "base_search_accuracy": 0.0625,
      "base_oracle_accuracy": 0.0625
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "dagger_policy",
      "split": "fresh_paired",
      "mode": "oracle_teacher",
      "k": 4,
      "n": 16,
      "accuracy": 0.4375,
      "valid_rate": 0.4375,
      "program_exact": 0.375,
      "stop_rate": 0.4375,
      "false_stop_rate": 0.0,
      "mean_steps": 2.875,
      "base_accuracy": 0.0625,
      "base_search_accuracy": 0.0625,
      "base_oracle_accuracy": 0.0625
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "dagger_policy",
      "split": "fresh_paired",
      "mode": "oracle_teacher",
      "k": 8,
      "n": 16,
      "accuracy": 0.5625,
      "valid_rate": 0.625,
      "program_exact": 0.375,
      "stop_rate": 0.5625,
      "false_stop_rate": 0.0,
      "mean_steps": 4.8125,
      "base_accuracy": 0.0625,
      "base_search_accuracy": 0.0625,
      "base_oracle_accuracy": 0.0625
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "dagger_policy",
      "split": "fresh_paired",
      "mode": "learned_stop",
      "k": 0,
      "n": 16,
      "accuracy": 0.0625,
      "valid_rate": 0.6875,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 0.0,
      "base_accuracy": 0.0625,
      "base_search_accuracy": 0.0625,
      "base_oracle_accuracy": 0.0625
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "dagger_policy",
      "split": "fresh_paired",
      "mode": "learned_stop",
      "k": 1,
      "n": 16,
      "accuracy": 0.0625,
      "valid_rate": 0.3125,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 1.0,
      "base_accuracy": 0.0625,
      "base_search_accuracy": 0.0625,
      "base_oracle_accuracy": 0.0625
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "dagger_policy",
      "split": "fresh_paired",
      "mode": "learned_stop",
      "k": 2,
      "n": 16,
      "accuracy": 0.0625,
      "valid_rate": 0.25,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 2.0,
      "base_accuracy": 0.0625,
      "base_search_accuracy": 0.0625,
      "base_oracle_accuracy": 0.0625
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "dagger_policy",
      "split": "fresh_paired",
      "mode": "learned_stop",
      "k": 4,
      "n": 16,
      "accuracy": 0.0625,
      "valid_rate": 0.25,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 4.0,
      "base_accuracy": 0.0625,
      "base_search_accuracy": 0.0625,
      "base_oracle_accuracy": 0.0625
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "dagger_policy",
      "split": "fresh_paired",
      "mode": "learned_stop",
      "k": 8,
      "n": 16,
      "accuracy": 0.0625,
      "valid_rate": 0.25,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 8.0,
      "base_accuracy": 0.0625,
      "base_search_accuracy": 0.0625,
      "base_oracle_accuracy": 0.0625
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "dagger_policy",
      "split": "fresh_paired",
      "mode": "learned_forced",
      "k": 0,
      "n": 16,
      "accuracy": 0.0625,
      "valid_rate": 0.6875,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 0.0,
      "base_accuracy": 0.0625,
      "base_search_accuracy": 0.0625,
      "base_oracle_accuracy": 0.0625
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "dagger_policy",
      "split": "fresh_paired",
      "mode": "learned_forced",
      "k": 1,
      "n": 16,
      "accuracy": 0.0625,
      "valid_rate": 0.3125,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 1.0,
      "base_accuracy": 0.0625,
      "base_search_accuracy": 0.0625,
      "base_oracle_accuracy": 0.0625
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "dagger_policy",
      "split": "fresh_paired",
      "mode": "learned_forced",
      "k": 2,
      "n": 16,
      "accuracy": 0.0625,
      "valid_rate": 0.3125,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 2.0,
      "base_accuracy": 0.0625,
      "base_search_accuracy": 0.0625,
      "base_oracle_accuracy": 0.0625
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "dagger_policy",
      "split": "fresh_paired",
      "mode": "learned_forced",
      "k": 4,
      "n": 16,
      "accuracy": 0.0625,
      "valid_rate": 0.25,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 4.0,
      "base_accuracy": 0.0625,
      "base_search_accuracy": 0.0625,
      "base_oracle_accuracy": 0.0625
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "dagger_policy",
      "split": "fresh_paired",
      "mode": "learned_forced",
      "k": 8,
      "n": 16,
      "accuracy": 0.0625,
      "valid_rate": 0.25,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 8.0,
      "base_accuracy": 0.0625,
      "base_search_accuracy": 0.0625,
      "base_oracle_accuracy": 0.0625
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "dagger_policy",
      "split": "hard_composition",
      "mode": "base",
      "k": 0,
      "n": 16,
      "accuracy": 0.0625,
      "valid_rate": 0.5625,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 0.0,
      "base_accuracy": 0.0625,
      "base_search_accuracy": 0.0625,
      "base_oracle_accuracy": 0.0625
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "dagger_policy",
      "split": "hard_composition",
      "mode": "oracle_teacher",
      "k": 0,
      "n": 16,
      "accuracy": 0.0625,
      "valid_rate": 0.5625,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 0.0,
      "base_accuracy": 0.0625,
      "base_search_accuracy": 0.0625,
      "base_oracle_accuracy": 0.0625
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "dagger_policy",
      "split": "hard_composition",
      "mode": "oracle_teacher",
      "k": 1,
      "n": 16,
      "accuracy": 0.0625,
      "valid_rate": 0.5,
      "program_exact": 0.0,
      "stop_rate": 0.0625,
      "false_stop_rate": 0.0,
      "mean_steps": 0.9375,
      "base_accuracy": 0.0625,
      "base_search_accuracy": 0.0625,
      "base_oracle_accuracy": 0.0625
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "dagger_policy",
      "split": "hard_composition",
      "mode": "oracle_teacher",
      "k": 2,
      "n": 16,
      "accuracy": 0.1875,
      "valid_rate": 0.3125,
      "program_exact": 0.125,
      "stop_rate": 0.0625,
      "false_stop_rate": 0.0,
      "mean_steps": 1.875,
      "base_accuracy": 0.0625,
      "base_search_accuracy": 0.0625,
      "base_oracle_accuracy": 0.0625
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "dagger_policy",
      "split": "hard_composition",
      "mode": "oracle_teacher",
      "k": 4,
      "n": 16,
      "accuracy": 0.1875,
      "valid_rate": 0.25,
      "program_exact": 0.125,
      "stop_rate": 0.1875,
      "false_stop_rate": 0.0,
      "mean_steps": 3.5,
      "base_accuracy": 0.0625,
      "base_search_accuracy": 0.0625,
      "base_oracle_accuracy": 0.0625
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "dagger_policy",
      "split": "hard_composition",
      "mode": "oracle_teacher",
      "k": 8,
      "n": 16,
      "accuracy": 0.3125,
      "valid_rate": 0.5,
      "program_exact": 0.25,
      "stop_rate": 0.1875,
      "false_stop_rate": 0.0,
      "mean_steps": 6.75,
      "base_accuracy": 0.0625,
      "base_search_accuracy": 0.0625,
      "base_oracle_accuracy": 0.0625
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "dagger_policy",
      "split": "hard_composition",
      "mode": "learned_stop",
      "k": 0,
      "n": 16,
      "accuracy": 0.0625,
      "valid_rate": 0.5625,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 0.0,
      "base_accuracy": 0.0625,
      "base_search_accuracy": 0.0625,
      "base_oracle_accuracy": 0.0625
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "dagger_policy",
      "split": "hard_composition",
      "mode": "learned_stop",
      "k": 1,
      "n": 16,
      "accuracy": 0.0625,
      "valid_rate": 0.4375,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 1.0,
      "base_accuracy": 0.0625,
      "base_search_accuracy": 0.0625,
      "base_oracle_accuracy": 0.0625
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "dagger_policy",
      "split": "hard_composition",
      "mode": "learned_stop",
      "k": 2,
      "n": 16,
      "accuracy": 0.0625,
      "valid_rate": 0.4375,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 2.0,
      "base_accuracy": 0.0625,
      "base_search_accuracy": 0.0625,
      "base_oracle_accuracy": 0.0625
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "dagger_policy",
      "split": "hard_composition",
      "mode": "learned_stop",
      "k": 4,
      "n": 16,
      "accuracy": 0.0625,
      "valid_rate": 0.4375,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 4.0,
      "base_accuracy": 0.0625,
      "base_search_accuracy": 0.0625,
      "base_oracle_accuracy": 0.0625
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "dagger_policy",
      "split": "hard_composition",
      "mode": "learned_stop",
      "k": 8,
      "n": 16,
      "accuracy": 0.0625,
      "valid_rate": 0.4375,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 8.0,
      "base_accuracy": 0.0625,
      "base_search_accuracy": 0.0625,
      "base_oracle_accuracy": 0.0625
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "dagger_policy",
      "split": "hard_composition",
      "mode": "learned_forced",
      "k": 0,
      "n": 16,
      "accuracy": 0.0625,
      "valid_rate": 0.5625,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 0.0,
      "base_accuracy": 0.0625,
      "base_search_accuracy": 0.0625,
      "base_oracle_accuracy": 0.0625
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "dagger_policy",
      "split": "hard_composition",
      "mode": "learned_forced",
      "k": 1,
      "n": 16,
      "accuracy": 0.0625,
      "valid_rate": 0.4375,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 1.0,
      "base_accuracy": 0.0625,
      "base_search_accuracy": 0.0625,
      "base_oracle_accuracy": 0.0625
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "dagger_policy",
      "split": "hard_composition",
      "mode": "learned_forced",
      "k": 2,
      "n": 16,
      "accuracy": 0.0625,
      "valid_rate": 0.4375,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 2.0,
      "base_accuracy": 0.0625,
      "base_search_accuracy": 0.0625,
      "base_oracle_accuracy": 0.0625
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "dagger_policy",
      "split": "hard_composition",
      "mode": "learned_forced",
      "k": 4,
      "n": 16,
      "accuracy": 0.0625,
      "valid_rate": 0.4375,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 4.0,
      "base_accuracy": 0.0625,
      "base_search_accuracy": 0.0625,
      "base_oracle_accuracy": 0.0625
    },
    {
      "run": "smoke_recurrent_vm_repair",
      "phase": "dagger_policy",
      "split": "hard_composition",
      "mode": "learned_forced",
      "k": 8,
      "n": 16,
      "accuracy": 0.0625,
      "valid_rate": 0.4375,
      "program_exact": 0.0,
      "stop_rate": 0.0,
      "false_stop_rate": 0.0,
      "mean_steps": 8.0,
      "base_accuracy": 0.0625,
      "base_search_accuracy": 0.0625,
      "base_oracle_accuracy": 0.0625
    }
  ]
}