{
  "schema_version": 1,
  "builder_plan": {
    "scenario": "You already need task-specific training and want to reduce its memory or compute cost.",
    "change": "Change only the training technique on a small controlled run after checking prerequisites.",
    "baseline": "Your current training recipe with the same model, data split, and evaluation.",
    "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": "Held-out quality, peak memory, training time, and full compute cost to the target quality.",
    "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": "2609.21039",
  "title": "Stiefel-AdamW: Geometry-Aware AdamW for Linear Factorization Blocks",
  "summary": "Stiefel-AdamW replaces AdamW specifically for the factor matrices of linear factorization blocks (LoRA adapters, low-rank layers, query-key products), constraining one factor to the Stiefel manifold while keeping AdamW's diagonal preconditioning in ambient space with only a tangent projection and retraction added. The claimed cost is minimal implementation and compute overhead over AdamW, with no extra memory of significance, fitting a single 24 GB GPU workflow. Failure modes include the unquantified gains not transferring to the reader's specific fine-tuning workload, retraction/projection bugs in custom implementations, and possible interaction issues with existing training stacks (gradient clipping, mixed precision, sharded optimizers) that the paper's benchmark setup may not cover.",
  "rationale": "The method is cheap to trial on LoRA fine-tuning within a single-GPU budget, but the abstract's unquantified 'consistent improvements' justify workload-specific validation before production adoption.",
  "technique": "Stiefel-AdamW optimizer for linear factorization blocks",
  "family": "treino_eficiente",
  "recommendation": "testar",
  "published": "2026-09-17",
  "as_of": "2026-10-18",
  "editorial_status": "indexed",
  "source_url": "https://arxiv.org/abs/2609.21039",
  "independent_implementations": 0,
  "total_implementations": 1,
  "stars": 0,
  "citations": 0,
  "claims": [
    {
      "claim_id": "claim-01",
      "statement": "Stiefel-AdamW replaces AdamW specifically for the factor matrices of linear factorization blocks (LoRA adapters, low-rank layers, query-key products), constraining one factor to the Stiefel manifold while keeping AdamW's diagonal preconditioning in ambient space with only a tangent projection and retraction added. The claimed cost is minimal implementation and compute overhead over AdamW, with no extra memory of significance, fitting a single 24 GB GPU workflow. Failure modes include the unquantified gains not transferring to the reader's specific fine-tuning workload, retraction/projection bugs in custom implementations, and possible interaction issues with existing training stacks (gradient clipping, mixed precision, sharded optimizers) that the paper's benchmark setup may not cover.",
      "basis": "inferred",
      "result": "",
      "baseline": "",
      "conditions": "",
      "source_url": "https://arxiv.org/abs/2609.21039",
      "source_page": null,
      "source_excerpt": ""
    },
    {
      "claim_id": "claim-02",
      "statement": "The abstract reports 'consistent improvements over strong baselines' on LoRA fine-tuning of GPT2, ViT, and Mistral 7B and GPT2 pretraining, but provides no quantified margin in the abstract.",
      "basis": "inferred",
      "result": "",
      "baseline": "",
      "conditions": "",
      "source_url": "https://arxiv.org/abs/2609.21039",
      "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": ""
}