{
  "schema_version": 1,
  "builder_plan": {
    "scenario": "Your self-hosted model spends too long generating answers token by token.",
    "change": "Check whether a compatible draft model and serving integration exist before renting compute.",
    "baseline": "The same target with speculation off, then your existing drafter if you have one.",
    "steps": [
      "Locate the paper's implementation and check its license, model support, data needs, and hardware requirements. Stop here if you cannot obtain the required artifacts.",
      "Save a small set of representative product tasks and expected outcomes. Keep evaluation tasks out of any training or tuning, and choose quality and cost limits before running.",
      "Run the baseline and the proposed change on identical inputs. Keep other settings fixed and repeat paired runs to expose variation.",
      "Save per-task outputs, failures, timings, and all usage costs. Review regressions before expanding the test."
    ],
    "measure": "Accepted draft length, tokens per second, tail latency, output quality, and total GPU cost.",
    "decision": "Keep the change only if it meets your preselected quality and latency limits and improves the metric you care about after setup and operating costs. Otherwise retain the baseline. This is an exploration plan, not a validated recommendation."
  },
  "builder_review": null,
  "arxiv_id": "2610.29210",
  "title": "Understanding Target Probability Reuse in Speculative Decoding",
  "summary": "It replaces draft-model ranking of the speculative token tree with reranking and first-token sampling based on a separately computed target-model distribution at the current context. It costs one additional target-model forward pass per round plus scheduling complexity to overlap that pass with drafting, and exact probability reuse may require discarding accepted tokens in unrestricted target models. It can fail if the extra target pass cannot be overlapped with drafting, since the added latency can erase the acceptance gains, and benefits shrink at large tree budgets.",
  "rationale": "The acceptance and latency benefits depend on implementation-specific overlap of the extra target pass and on tree-budget regime, so workload-specific validation on the reader's single-GPU or API setup is required before production adoption.",
  "technique": "Target-probability-guided tree allocation in speculative decoding",
  "family": "decodificacao_especulativa",
  "recommendation": "testar",
  "published": "2026-10-21",
  "as_of": "2026-10-22",
  "editorial_status": "indexed",
  "source_url": "https://arxiv.org/abs/2610.29210",
  "independent_implementations": 2,
  "total_implementations": 3,
  "stars": 126,
  "citations": null,
  "claims": [
    {
      "claim_id": "claim-01",
      "statement": "It replaces draft-model ranking of the speculative token tree with reranking and first-token sampling based on a separately computed target-model distribution at the current context. It costs one additional target-model forward pass per round plus scheduling complexity to overlap that pass with drafting, and exact probability reuse may require discarding accepted tokens in unrestricted target models. It can fail if the extra target pass cannot be overlapped with drafting, since the added latency can erase the acceptance gains, and benefits shrink at large tree budgets.",
      "basis": "inferred",
      "result": "",
      "baseline": "",
      "conditions": "",
      "source_url": "https://arxiv.org/abs/2610.29210",
      "source_page": null,
      "source_excerpt": ""
    },
    {
      "claim_id": "claim-02",
      "statement": "Both selection rules decode faster than draft ranking when the separate target distribution pass overlaps standalone drafting, and improve expected acceptance over draft-based ranking, especially with small tree budgets; no numeric speedup factor is reported in the abstract.",
      "basis": "inferred",
      "result": "",
      "baseline": "",
      "conditions": "",
      "source_url": "https://arxiv.org/abs/2610.29210",
      "source_page": null,
      "source_excerpt": ""
    }
  ],
  "exposure_map": [
    {
      "dimension": "quality",
      "basis": "not_evaluated",
      "finding": "",
      "interpretation": "",
      "limitation": "",
      "next_check": "",
      "source_url": "",
      "source_page": null,
      "source_excerpt": ""
    },
    {
      "dimension": "compute",
      "basis": "not_evaluated",
      "finding": "",
      "interpretation": "",
      "limitation": "",
      "next_check": "",
      "source_url": "",
      "source_page": null,
      "source_excerpt": ""
    },
    {
      "dimension": "latency",
      "basis": "not_evaluated",
      "finding": "",
      "interpretation": "",
      "limitation": "",
      "next_check": "",
      "source_url": "",
      "source_page": null,
      "source_excerpt": ""
    },
    {
      "dimension": "operations",
      "basis": "not_evaluated",
      "finding": "",
      "interpretation": "",
      "limitation": "",
      "next_check": "",
      "source_url": "",
      "source_page": null,
      "source_excerpt": ""
    },
    {
      "dimension": "compatibility",
      "basis": "not_evaluated",
      "finding": "",
      "interpretation": "",
      "limitation": "",
      "next_check": "",
      "source_url": "",
      "source_page": null,
      "source_excerpt": ""
    },
    {
      "dimension": "security",
      "basis": "not_evaluated",
      "finding": "",
      "interpretation": "",
      "limitation": "",
      "next_check": "",
      "source_url": "",
      "source_page": null,
      "source_excerpt": ""
    },
    {
      "dimension": "data_and_training",
      "basis": "not_evaluated",
      "finding": "",
      "interpretation": "",
      "limitation": "",
      "next_check": "",
      "source_url": "",
      "source_page": null,
      "source_excerpt": ""
    },
    {
      "dimension": "reproducibility",
      "basis": "not_evaluated",
      "finding": "",
      "interpretation": "",
      "limitation": "",
      "next_check": "",
      "source_url": "",
      "source_page": null,
      "source_excerpt": ""
    }
  ],
  "risks": [],
  "minimum_test": [],
  "open_questions": [],
  "independent_tests": [],
  "lab_reproductions": [],
  "report_available": false,
  "validation_tier": "unknown",
  "evidence_tier": "unknown",
  "infrastructure_basis": "unknown",
  "equations": [],
  "equations_status": "not_fetched",
  "core_kind": "",
  "equations_fetched_at": "",
  "equations_source_url": "",
  "equations_source_sha256": "",
  "exposures_reviewed_at": "",
  "exposures_source_sha256": ""
}