{
  "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": "2609.24847",
  "title": "SPECTRA: Adaptive Execution of Speculative Decoding on a Runtime-Reconfigurable Tiled Architecture",
  "summary": "Replaces fixed systolic or fixed vector execution engines in a custom tiled accelerator with per-kernel runtime switching between systolic (GEMM) and vector-lane (GEMV) modes, plus dynamic tile-count, partitioning, and communication reconfiguration for speculative decoding's variable arithmetic intensity. Costs include designing or sourcing a reconfigurable tiled hardware architecture, an FPGA prototype platform, and a runtime controller for per-kernel reconfiguration; it does not run on commodity GPUs. It can fail if acceptance rates or speculation lengths shift the regime in ways the reconfiguration granularity cannot track, and the reported gains are relative to fixed variants of the same prototype rather than to GPU baselines.",
  "rationale": "The contribution is a custom FPGA hardware architecture, which the reader cannot deploy with a single 24 GB GPU or third-party APIs, though the underlying insight about speculative decoding's variable arithmetic intensity is worth noting when configuring speculative decoding on existing GPU stacks.",
  "technique": "Runtime-reconfigurable tiled FPGA for speculative decoding",
  "family": "decodificacao_especulativa",
  "recommendation": "nao_aplica",
  "published": "2026-09-21",
  "as_of": "2026-10-18",
  "editorial_status": "indexed",
  "source_url": "https://arxiv.org/abs/2609.24847",
  "independent_implementations": 0,
  "total_implementations": 0,
  "stars": 0,
  "citations": 0,
  "claims": [
    {
      "claim_id": "claim-01",
      "statement": "Replaces fixed systolic or fixed vector execution engines in a custom tiled accelerator with per-kernel runtime switching between systolic (GEMM) and vector-lane (GEMV) modes, plus dynamic tile-count, partitioning, and communication reconfiguration for speculative decoding's variable arithmetic intensity. Costs include designing or sourcing a reconfigurable tiled hardware architecture, an FPGA prototype platform, and a runtime controller for per-kernel reconfiguration; it does not run on commodity GPUs. It can fail if acceptance rates or speculation lengths shift the regime in ways the reconfiguration granularity cannot track, and the reported gains are relative to fixed variants of the same prototype rather than to GPU baselines.",
      "basis": "inferred",
      "result": "",
      "baseline": "",
      "conditions": "",
      "source_url": "https://arxiv.org/abs/2609.24847",
      "source_page": null,
      "source_excerpt": ""
    },
    {
      "claim_id": "claim-02",
      "statement": "Up to 2.09x speedup from tile-level reconfiguration and a further 1.25x from system-level adaptability over fixed designs, on a 20-tile FPGA prototype across Pythia, SmolLM2, and GPT-2 families.",
      "basis": "inferred",
      "result": "",
      "baseline": "",
      "conditions": "",
      "source_url": "https://arxiv.org/abs/2609.24847",
      "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": ""
}