{
  "method": "supervised warm start + on-policy contextual-bandit REINFORCE",
  "task": "One-step routing reward from authored labels; not end-to-end factuality or real-user feedback.",
  "seed": 20260909,
  "dataset_sha256": "965244ecb66e21f43b52a66ac0b5d47781c158a49c3ac86c825b050d374ada10",
  "training_seconds": 3.101,
  "max_training_seconds": 120,
  "rl_updates": 1620,
  "parameters": 13079,
  "split_counts": {
    "train": 1830,
    "validation": 420,
    "test": 438
  },
  "split_protocol": "Topics are disjoint across all three splits. Training question templates differ from validation/test templates. Authored cases; no user conversations.",
  "old_policy_test": {
    "mode": 0.9429223744292238,
    "intent": 0.9928571428571429,
    "topic_count": 0.1761904761904762,
    "frag_count": 0,
    "tone": 0.5095238095238095,
    "joint_mode_intent": 0.9360730593607306,
    "score": 0.7919308545335944
  },
  "supervised_test": {
    "mode": 0.9885844748858448,
    "intent": 0.9976190476190476,
    "topic_count": 1,
    "frag_count": 1,
    "tone": 1,
    "joint_mode_intent": 0.9863013698630136,
    "score": 0.9944814090019569
  },
  "reinforce_test": {
    "mode": 0.9885844748858448,
    "intent": 0.9976190476190476,
    "topic_count": 1,
    "frag_count": 1,
    "tone": 1,
    "joint_mode_intent": 0.9863013698630136,
    "score": 0.9944814090019569
  },
  "validation": {
    "mode": 1,
    "intent": 0.9926470588235294,
    "topic_count": 1,
    "frag_count": 0.9926470588235294,
    "tone": 0.9926470588235294,
    "joint_mode_intent": 0.9928571428571429,
    "score": 0.9963235294117647
  },
  "release_gate_passed": true,
  "retrieval_benchmark": {
    "runtime": "v22.22.3",
    "corpus_entries": 80,
    "queries": 30,
    "samples": 150,
    "old": {
      "median_ms": 16.666615000000093,
      "p95_ms": 17.66905799999995
    },
    "current": {
      "median_ms": 0.07340399999975489,
      "p95_ms": 0.1297669999999016
    },
    "median_speedup": 227.05322598299475,
    "ranking_parity": true
  },
  "numerical_verification": {
    "fixtures": 40,
    "maxExportError": 1.7881393432617188e-7,
    "maxWasmError": 0,
    "js": {
      "median_us": 13.968000000005532,
      "p95_us": 18.507999999997082
    },
    "wasm": {
      "median_us": 14.24799999998072,
      "p95_us": 21.721000000013646
    },
    "wasmBytes": 3685,
    "node": "v22.22.3"
  }
}