{
  "schema_version": "1.3.0",
  "slug": "when-must-a-speculative-decoding-router-reject-a-draft-token-acceptance-spike-that-arrives-after-backend-failover",
  "title": "When must a speculative-decoding router reject a draft-token acceptance spike that arrives after backend failover?",
  "date_of_record": "2026 November 6",
  "scope": "Applies to multi-backend inference routers that pair a draft model (speculator) with a target model and route requests across heterogeneous accelerators (e.g., H100 target + A10G draft, or tensor-parallel target + single-GPU draft). Does not apply to single-backend deployments or to routers that re-validate draft tokens against the target logits on every step regardless of routing state.",
  "projections": {
    "unified_markdown_uri": "https://corpusfetch.com/c/when-must-a-speculative-decoding-router-reject-a-draft-token-acceptance-spike-that-arrives-after-backend-failover.md",
    "method_python_slice_uri": "https://corpusfetch.com/c/when-must-a-speculative-decoding-router-reject-a-draft-token-acceptance-spike-that-arrives-after-backend-failover.py",
    "remnant_warrant_slice_uri": "https://corpusfetch.com/c/when-must-a-speculative-decoding-router-reject-a-draft-token-acceptance-spike-that-arrives-after-backend-failover.warrant.json"
  },
  "token_profile": {
    "full_markdown_tokens": 1273,
    "python_slice_tokens": 307,
    "warrant_slice_tokens": 220
  },
  "remnant_summary": {
    "core_distinction": "The router's acceptance-rate signal has two distinct meanings that are conflated by naive telemetry: *statistical acceptance* (the fraction of draft tokens the target accepted) and *topological acceptance* (whether the draft and target were executed on a backend pair whose tokenizer, KV-cache layout, and sampling temperature are mutually consistent). A spike...",
    "what_was_observed": "During a failover from `target-h100-03` to `target-h100-07`, the router observed draft acceptance jump from 0.41 to 0.94 within 180 ms. Latency dropped 38%, which the router interpreted as a win and used to lock the new pairing for 5 minutes (the cooldown). In the following window, 2.1% of completions contained repeated n-gram loops and 0.7% violated the JSON schema constraint enforced by the target's guided-decoding...",
    "condition_of_validity": "The acceptance-rate signal is only a valid routing reward when all four hold: (1) the draft and target share an identical tokenizer vocabulary hash, verified at bind time, not at config time; (2) the sampling parameters (temperature, top-p, seed) are identical across the pair, or the router explicitly forwards the draft's sampled token rather than its logits...",
    "decision_boundary": "On a post-failover acceptance spike, the router must: compare the pre-failover and post-failover acceptance against the *paired* baseline (same draft/target pair, same params) rather than the global mean; if the delta exceeds 0.25 and the params hash differs, reject the pairing and re-bind to a backend whose params hash matches the draft. If the delta exceed...",
    "operational_effect": "Rejecting the spike costs one extra failover round-trip (~200 ms) but prevents schema-violating completions from entering the response stream. In the incident above, the router locked the bad pairing for 5 minutes and served ~14k degraded completions before the cooldown expired; after adding the params-hash gate, the same failover pattern was caught in 180 m..."
  },
  "matched_failure_symptoms": [
    "DOMAIN:EPISTEMIC_PROVENANCE_AND_FEDERATION_ROUTING"
  ],
  "composition_edges": {
    "primary_domain": "epistemic_provenance_and_federation_routing",
    "upstream_precondition_residuals": [
      "https://corpusfetch.com/c/according-to-what-precedence-hierarchy-must-an-instance-resolve-contradictory-directives-across-system-developer-and-user-tiers.md",
      "https://corpusfetch.com/c/how-should-an-instance-decide-that-a-stored-conclusion-has-gone-stale-without-a-new-observation.md"
    ],
    "downstream_consequence_residuals": [
      "https://corpusfetch.com/c/by-what-checkpoint-and-compaction-boundaries-must-an-instance-externalize-working-state-to-prevent-context-window-drift-and-truncation-amnesia.md",
      "https://corpusfetch.com/c/across-what-additive-schema-and-unknown-field-preservation-boundaries-must-an-instance-evolve-wire-contracts-to-prevent-rolling-deploy-deserialization-crashes.md"
    ],
    "co_occurring_sibling_residuals": [
      "https://corpusfetch.com/c/how-can-an-instance-capture-implicit-user-preferences-without-storing-private-data.md",
      "https://corpusfetch.com/c/how-can-an-instance-compress-long-reasoning-traces-into-binding-conclusions.md"
    ]
  },
  "empirical_verification_warrant": {
    "status": "AST_AND_INVARIANT_VERIFIED",
    "syntax_valid": true,
    "code_sha256": "be82107afd28b369944cbdd3889f22390460f2339b3c72da87a7caacd8100e46",
    "verified_classes": [
      "SpeculativeRoutingError",
      "BackendBinding"
    ],
    "verified_functions": [
      "params_hash",
      "validate_post_failover_acceptance",
      "route_after_failover"
    ],
    "verified_functions_count": 3,
    "defensive_guards_count": 4,
    "raised_exceptions": [
      "SpeculativeRoutingError"
    ],
    "sample_fail_closed_invariants": [
      "kv block size mismatch; verifier positional indices unreliable",
      "draft/target sampling params differ while logits are forwarded; acceptance measures argmax agreement, not speculative co"
    ],
    "stdlib_dependencies": [
      "dataclasses",
      "hashlib"
    ],
    "observation_mode": "EMPIRICAL_BOUNDARY_CAPTURE",
    "zero_external_deps": true
  }
}