{
  "schema_version": 1,
  "builder_plan": {
    "scenario": "A support or data assistant repeats work across related customer requests.",
    "change": "Try the paper's memory change in a replay of past tasks, with separate memory per customer.",
    "baseline": "Your current memory pipeline plus a no-memory control using the same agent.",
    "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 success, stale memories, retrieval misses, response latency, and all memory maintenance costs.",
    "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": "2609.35578",
  "title": "FactorEngram: Factorized N-gram Memory with Basis-Level Gating for Language Models",
  "summary": "The method replaces monolithic hashed n-gram embedding lookups (as in Engram) with a shared dictionary of basis vectors, sparsity-regularized coefficients, and per-basis contextual gating scored against the backbone hidden state. It costs additional memory-branch parameters, per-basis scoring computation, and training complexity, and it requires training the backbone jointly with the memory rather than retrofitting an existing model. Adoption can fail because benefits are demonstrated only on 340M and 1B models trained with the branch in place, insertion depth is sensitive (middle layers before attention were found effective), and hash-table and dictionary sizing introduce new hyperparameters whose gains may not transfer to other scales or workloads.",
  "rationale": "The technique requires training a backbone jointly with the memory branch, which is infeasible under the reader's constrained infrastructure, and the abstract offers no quantified gains at a scale the reader could reproduce.",
  "technique": "Factorized n-gram memory with basis-level contextual gating",
  "family": "memoria_e_contexto",
  "recommendation": "observar",
  "published": "2026-09-28",
  "as_of": "2026-10-18",
  "editorial_status": "indexed",
  "source_url": "https://arxiv.org/abs/2609.35578",
  "independent_implementations": 0,
  "total_implementations": 1,
  "stars": 0,
  "citations": null,
  "claims": [
    {
      "claim_id": "claim-01",
      "statement": "The method replaces monolithic hashed n-gram embedding lookups (as in Engram) with a shared dictionary of basis vectors, sparsity-regularized coefficients, and per-basis contextual gating scored against the backbone hidden state. It costs additional memory-branch parameters, per-basis scoring computation, and training complexity, and it requires training the backbone jointly with the memory rather than retrofitting an existing model. Adoption can fail because benefits are demonstrated only on 340M and 1B models trained with the branch in place, insertion depth is sensitive (middle layers before attention were found effective), and hash-table and dictionary sizing introduce new hyperparameters whose gains may not transfer to other scales or workloads.",
      "basis": "inferred",
      "result": "",
      "baseline": "",
      "conditions": "",
      "source_url": "https://arxiv.org/abs/2609.35578",
      "source_page": null,
      "source_excerpt": ""
    },
    {
      "claim_id": "claim-02",
      "statement": "Improves language modeling and downstream task performance on 340M- and 1B-parameter backbones; no quantified figures are given in the abstract.",
      "basis": "inferred",
      "result": "",
      "baseline": "",
      "conditions": "",
      "source_url": "https://arxiv.org/abs/2609.35578",
      "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": ""
}