{
  "schema_version": 1,
  "builder_plan": {
    "scenario": "A self-hosted model uses more compute or memory than your workload can justify.",
    "change": "Check for a usable pruned checkpoint and runtime support before attempting model surgery.",
    "baseline": "The original unpruned model at the same precision, context, and load.",
    "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": "Task quality, actual GPU memory, serving throughput, and preparation 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": "2610.29662",
  "title": "SPARE: Sparsity Allocation via Routing Entropy for Efficient MoEs",
  "summary": "SPARE replaces uniform per-layer expert pruning budgets with a layer-wise allocation derived from routing entropy measured on calibration data, applied on top of any existing expert selector. It is training-free, so its cost is a calibration forward pass and the engineering of pruning and remapping experts; the pruned model is smaller in memory but loses capacity relative to the dense model. It can fail when calibration data is unrepresentative of production traffic, since routing entropy estimates then misallocate the pruning budget, and gains shrink at low pruning ratios where uniform allocation is already adequate.",
  "rationale": "The method is training-free and selector-agnostic, so a small team can validate it on an existing MoE checkpoint with modest calibration compute, but its benefit over uniform pruning is workload- and ratio-dependent and must be confirmed on the reader's own tasks.",
  "technique": "Entropy-guided per-layer MoE expert pruning",
  "family": "esparsidade_e_poda",
  "recommendation": "testar",
  "published": "2026-10-21",
  "as_of": "2026-10-22",
  "editorial_status": "indexed",
  "source_url": "https://arxiv.org/abs/2610.29662",
  "independent_implementations": 0,
  "total_implementations": 0,
  "stars": 0,
  "citations": null,
  "claims": [
    {
      "claim_id": "claim-01",
      "statement": "SPARE replaces uniform per-layer expert pruning budgets with a layer-wise allocation derived from routing entropy measured on calibration data, applied on top of any existing expert selector. It is training-free, so its cost is a calibration forward pass and the engineering of pruning and remapping experts; the pruned model is smaller in memory but loses capacity relative to the dense model. It can fail when calibration data is unrepresentative of production traffic, since routing entropy estimates then misallocate the pruning budget, and gains shrink at low pruning ratios where uniform allocation is already adequate.",
      "basis": "inferred",
      "result": "",
      "baseline": "",
      "conditions": "",
      "source_url": "https://arxiv.org/abs/2610.29662",
      "source_page": null,
      "source_excerpt": ""
    },
    {
      "claim_id": "claim-02",
      "statement": "At 50% expert pruning, SPARE raises the eight-task average of Qwen3-30B-A3B from 57.2% to 63.8% versus uniform allocation, and the math/code/science average of DeepSeek-R1 from 80.1 to 82.8.",
      "basis": "inferred",
      "result": "",
      "baseline": "",
      "conditions": "",
      "source_url": "https://arxiv.org/abs/2610.29662",
      "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": ""
}