{ "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": "2609.27980", "title": "Six Layers Less: Encoder Pruning for Whisper with Label-Free Recovery", "summary": "Replaces the full 32-layer Whisper large-v3-turbo encoder with a 26-layer encoder selected by leave-one-layer-out WER ranking, recovered via distillation on unlabeled monolingual speech. The cost is a mean WER increase of roughly 1.9 points (18.2% to 20.1%) plus a distillation run requiring unlabeled speech data and modest GPU time; inference itself needs no custom code. What can fail: the ~10% relative WER regression may be unacceptable for high-accuracy domains, the layer ranking and recovery may not transfer to languages or acoustic conditions outside the four evaluated, and realized latency gains depend on whether the encoder is the bottleneck in the deployment stack.", "rationale": "The method is deployable on a single 24 GB GPU with no custom inference code and a released model, but the measured WER regression and limited language coverage require validation on the reader's own audio distribution before production adoption.", "technique": "Whisper encoder layer pruning with label-free distillation", "family": "esparsidade_e_poda", "recommendation": "testar", "published": "2026-09-23", "as_of": "2026-10-18", "editorial_status": "indexed", "source_url": "https://arxiv.org/abs/2609.27980", "independent_implementations": 2, "total_implementations": 3, "stars": 4, "citations": 0, "claims": [ { "claim_id": "claim-01", "statement": "Replaces the full 32-layer Whisper large-v3-turbo encoder with a 26-layer encoder selected by leave-one-layer-out WER ranking, recovered via distillation on unlabeled monolingual speech. The cost is a mean WER increase of roughly 1.9 points (18.2% to 20.1%) plus a distillation run requiring unlabeled speech data and modest GPU time; inference itself needs no custom code. What can fail: the ~10% relative WER regression may be unacceptable for high-accuracy domains, the layer ranking and recovery may not transfer to languages or acoustic conditions outside the four evaluated, and realized latency gains depend on whether the encoder is the bottleneck in the deployment stack.", "basis": "inferred", "result": "", "baseline": "", "conditions": "", "source_url": "https://arxiv.org/abs/2609.27980", "source_page": null, "source_excerpt": "" }, { "claim_id": "claim-02", "statement": "Removes 6 of 32 encoder layers (18.5% of the encoder stack) from whisper-large-v3-turbo with no custom inference code; mean WER across four languages rises from 18.2% baseline to 20.1% after unlabeled distillation (21.9% zero-shot).", "basis": "inferred", "result": "", "baseline": "", "conditions": "", "source_url": "https://arxiv.org/abs/2609.27980", "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": "" }