{
  "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": "2609.19868",
  "title": "Zarya: A Hybrid Autoregressive--Masked Diffusion Language Model with Flexible Training and Dual-Mode Inference",
  "summary": "Zarya replaces a purely autoregressive decoding stack with a single model trained jointly on AR and masked-diffusion objectives, offering either parallel MDM sampling or slotted speculative decoding that interleaves diffusion-based slot selection with AR infilling while reusing the KV cache. The cost is a custom training curriculum (variable-size slots with increasing granularity) and a non-standard inference interface; pure MDM mode still forfeits KV-cache reuse and incurs its characteristic compute overhead, and no off-the-shelf serving stack supports either decoding mode. Failures after adoption include incoherent generation in diffusion mode, immature tooling and reproducibility risk for the speculative mode, and unverified speedups, since the abstract reports benchmark participation but no quantified latency or quality advantage over a standard AR baseline at the released 0.6B–4B scales.",
  "rationale": "The released model sizes fit a 24 GB GPU and the KV-cache-reusing speculative mode is directly relevant to constrained inference, but the absence of quantified gains and the need for a custom decoding stack mean adoption should follow small-scale latency and quality validation on the reader's workload.",
  "technique": "Hybrid AR–masked diffusion LM with dual decoding",
  "family": "decodificacao_especulativa",
  "recommendation": "testar",
  "published": "2026-09-17",
  "as_of": "2026-10-18",
  "editorial_status": "indexed",
  "source_url": "https://arxiv.org/abs/2609.19868",
  "independent_implementations": 0,
  "total_implementations": 1,
  "stars": 2,
  "citations": 0,
  "claims": [
    {
      "claim_id": "claim-01",
      "statement": "Zarya replaces a purely autoregressive decoding stack with a single model trained jointly on AR and masked-diffusion objectives, offering either parallel MDM sampling or slotted speculative decoding that interleaves diffusion-based slot selection with AR infilling while reusing the KV cache. The cost is a custom training curriculum (variable-size slots with increasing granularity) and a non-standard inference interface; pure MDM mode still forfeits KV-cache reuse and incurs its characteristic compute overhead, and no off-the-shelf serving stack supports either decoding mode. Failures after adoption include incoherent generation in diffusion mode, immature tooling and reproducibility risk for the speculative mode, and unverified speedups, since the abstract reports benchmark participation but no quantified latency or quality advantage over a standard AR baseline at the released 0.6B–4B scales.",
      "basis": "inferred",
      "result": "",
      "baseline": "",
      "conditions": "",
      "source_url": "https://arxiv.org/abs/2609.19868",
      "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": ""
}