{ "schema_version": 1, "builder_plan": { "scenario": "Your workload might fit a smaller or differently structured model.", "change": "Check for available weights and a compatible runtime before planning a product trial.", "baseline": "Your current model on identical held-out tasks and the same resource budget.", "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, latency, memory, context limits, and migration 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": "2609.27735", "title": "NS-ATTENTION: Newton-Schulz Transformations of Attention Outputs in Vision Transformers", "summary": "The method inserts a parameter-free Newton-Schulz polynomial step on each attention head's output matrix (after Frobenius normalization, with the norm restored afterward) between attention computation and the standard head merge and output projection, replacing nothing but augmenting that path. It adds inference latency and extra per-head matrix polynomial computation at both training and inference, with no parameters or memory overhead beyond transient buffers. Gains are demonstrated only on small-scale image classification (ViT/Swin on CIFAR), so the effect may not transfer to other modalities, larger models, or pre-trained checkpoints, and the one-versus-two-iteration ablation suggests sensitivity to the iteration count.", "rationale": "The technique is cheap to implement and parameter-free, but its benefits are validated only on small vision benchmarks with modest accuracy gains and a latency cost, so it warrants a small-scale trial on the reader's own workload rather than direct adoption.", "technique": "Newton-Schulz transform of attention head outputs", "family": "arquitetura_eficiente", "recommendation": "testar", "published": "2026-09-23", "as_of": "2026-10-18", "editorial_status": "indexed", "source_url": "https://arxiv.org/abs/2609.27735", "independent_implementations": 0, "total_implementations": 0, "stars": 0, "citations": 0, "claims": [ { "claim_id": "claim-01", "statement": "The method inserts a parameter-free Newton-Schulz polynomial step on each attention head's output matrix (after Frobenius normalization, with the norm restored afterward) between attention computation and the standard head merge and output projection, replacing nothing but augmenting that path. It adds inference latency and extra per-head matrix polynomial computation at both training and inference, with no parameters or memory overhead beyond transient buffers. Gains are demonstrated only on small-scale image classification (ViT/Swin on CIFAR), so the effect may not transfer to other modalities, larger models, or pre-trained checkpoints, and the one-versus-two-iteration ablation suggests sensitivity to the iteration count.", "basis": "inferred", "result": "", "baseline": "", "conditions": "", "source_url": "https://arxiv.org/abs/2609.27735", "source_page": null, "source_excerpt": "" }, { "claim_id": "claim-02", "statement": "Mean accuracy gains of 0.25-0.83 percentage points over baseline across 12 matched-seed comparisons of ViT and Swin on CIFAR-10/100, with reduced leading-eigenvalue concentration and increased effective rank.", "basis": "inferred", "result": "", "baseline": "", "conditions": "", "source_url": "https://arxiv.org/abs/2609.27735", "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": "" }