{
  "schema_version": 1,
  "builder_plan": {
    "scenario": "A product assistant retrieves plausible documents but gives unsupported answers.",
    "change": "Change one retrieval or ranking step while keeping the document snapshot fixed.",
    "baseline": "Your current retrieval pipeline with identical questions and answer model.",
    "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": "Supported answers, retrieval misses, correct abstentions, index cost, and response latency.",
    "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": "2610.08732",
  "title": "A Systematic Study of Semantic ID Spaces for Generative Information Retrieval",
  "summary": "The work replaces expensive end-to-end model training as the evaluation loop for DocID design with a suite of training-free intrinsic metrics, plus a unified PQ/RQ framework for comparing DocID structures. The cost is adopting the generative retrieval paradigm itself: training or fine-tuning a seq2seq model on document identifiers, maintaining codebooks, and handling identifier generation and constrained decoding at inference. What can fail is that intrinsic metric scores may not predict downstream retrieval quality on the reader's specific corpus, and GIR scales poorly to large or frequently updated document collections compared to conventional ANN indexes.",
  "rationale": "The intrinsic metrics are cheap to apply and genuinely useful for anyone building generative retrieval, but the entire approach only matters if the reader's workload justifies GIR over standard dense retrieval, which requires workload-specific validation on a 24 GB GPU budget.",
  "technique": "Intrinsic metrics for semantic DocID design in GIR",
  "family": "recuperacao_e_rag",
  "recommendation": "testar",
  "published": "2026-10-06",
  "as_of": "2026-10-18",
  "editorial_status": "indexed",
  "source_url": "https://arxiv.org/abs/2610.08732",
  "independent_implementations": 0,
  "total_implementations": 1,
  "stars": 22,
  "citations": null,
  "claims": [
    {
      "claim_id": "claim-01",
      "statement": "The work replaces expensive end-to-end model training as the evaluation loop for DocID design with a suite of training-free intrinsic metrics, plus a unified PQ/RQ framework for comparing DocID structures. The cost is adopting the generative retrieval paradigm itself: training or fine-tuning a seq2seq model on document identifiers, maintaining codebooks, and handling identifier generation and constrained decoding at inference. What can fail is that intrinsic metric scores may not predict downstream retrieval quality on the reader's specific corpus, and GIR scales poorly to large or frequently updated document collections compared to conventional ANN indexes.",
      "basis": "inferred",
      "result": "",
      "baseline": "",
      "conditions": "",
      "source_url": "https://arxiv.org/abs/2610.08732",
      "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": ""
}