{ "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": "2610.01554", "title": "QK-Wanda: Coupling Queries and Keys for Unstructured Pruning", "summary": "QK-Wanda replaces Wanda's per-projection independent weight scoring with closed-form scores that couple query and key projections under a shared pruning budget, using an unmasked pre-RoPE reconstruction objective. It costs only 1.3-3.1% more pruning time than Wanda, requires no gradients or weight updates, and preserves Wanda's one-shot calibration workflow. It can fail because lower local reconstruction error does not reliably predict downstream quality, as shown by Llama-3.1-70B's perplexity regression, and 80% unstructured sparsity yields limited practical speedup without sparse kernels.", "rationale": "It is a near-drop-in improvement over Wanda with negligible overhead and meaningful quality gains on some models, but the demonstrated model-dependent regressions require validation on the reader's specific model before production adoption.", "technique": "Coupled QK scoring for unstructured pruning", "family": "esparsidade_e_poda", "recommendation": "testar", "published": "2026-10-01", "as_of": "2026-10-18", "editorial_status": "indexed", "source_url": "https://arxiv.org/abs/2610.01554", "independent_implementations": 1, "total_implementations": 2, "stars": 4, "citations": null, "claims": [ { "claim_id": "claim-01", "statement": "QK-Wanda replaces Wanda's per-projection independent weight scoring with closed-form scores that couple query and key projections under a shared pruning budget, using an unmasked pre-RoPE reconstruction objective. It costs only 1.3-3.1% more pruning time than Wanda, requires no gradients or weight updates, and preserves Wanda's one-shot calibration workflow. It can fail because lower local reconstruction error does not reliably predict downstream quality, as shown by Llama-3.1-70B's perplexity regression, and 80% unstructured sparsity yields limited practical speedup without sparse kernels.", "basis": "inferred", "result": "", "baseline": "", "conditions": "", "source_url": "https://arxiv.org/abs/2610.01554", "source_page": null, "source_excerpt": "" }, { "claim_id": "claim-02", "statement": "At 80% sparsity on Llama 2 70B, WikiText-2 and C4 perplexity decrease by 20.3% and 13.5% versus Wanda, with mean zero-shot accuracy up 5.94 points; QK reconstruction error falls 60% at 50% and 45% at 80% sparsity on average, but Llama-3.1-70B shows substantially higher perplexity despite lower reconstruction error.", "basis": "inferred", "result": "", "baseline": "", "conditions": "", "source_url": "https://arxiv.org/abs/2610.01554", "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": "" }