{
  "schema_version": 1,
  "builder_plan": {
    "scenario": "A self-hosted assistant is too expensive because its model uses most of the GPU memory.",
    "change": "Try the paper's supported lower-precision weights on a copy of your serving setup.",
    "baseline": "The same model at its original precision, on the same GPU.",
    "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": "Peak memory, task pass rate, latency, and total serving cost at the same context and load.",
    "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": "2609.28270",
  "title": "Predicting Quantization Price for Selecting PTQ Configurations Before Deployment",
  "summary": "It replaces ad hoc or separately scored PTQ configuration choices (format, granularity, quantizer family, transformations, bits) with a single pre-deployment price, computed as the trace of each layer's induced error covariance against estimated downstream Hessian curvature, enabling budgeted selection from a calibration-time price table. It costs one calibration pass with Hessian estimation and covariance evaluation per candidate configuration, adding offline compute and implementation complexity but no deployment overhead. It can fail if the estimated curvature is a poor proxy for true output-distribution drift, if the second-order KL approximation breaks down at very low bit widths, or if candidate covariances are misestimated from limited calibration data.",
  "rationale": "The method addresses a real production need (choosing PTQ configurations before deployment) at a cost compatible with a single 24 GB GPU, but the abstract provides no empirical validation, so the reader should validate its predictions against actual quantized-model quality on their own workloads before relying on it.",
  "technique": "Curvature-priced PTQ configuration selection",
  "family": "quantizacao",
  "recommendation": "testar",
  "published": "2026-09-23",
  "as_of": "2026-10-18",
  "editorial_status": "indexed",
  "source_url": "https://arxiv.org/abs/2609.28270",
  "independent_implementations": 0,
  "total_implementations": 0,
  "stars": 0,
  "citations": 0,
  "claims": [
    {
      "claim_id": "claim-01",
      "statement": "It replaces ad hoc or separately scored PTQ configuration choices (format, granularity, quantizer family, transformations, bits) with a single pre-deployment price, computed as the trace of each layer's induced error covariance against estimated downstream Hessian curvature, enabling budgeted selection from a calibration-time price table. It costs one calibration pass with Hessian estimation and covariance evaluation per candidate configuration, adding offline compute and implementation complexity but no deployment overhead. It can fail if the estimated curvature is a poor proxy for true output-distribution drift, if the second-order KL approximation breaks down at very low bit widths, or if candidate covariances are misestimated from limited calibration data.",
      "basis": "inferred",
      "result": "",
      "baseline": "",
      "conditions": "",
      "source_url": "https://arxiv.org/abs/2609.28270",
      "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": ""
}