{ "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": "2610.20588", "title": "Your Transformer Is Secretly Elastic", "summary": "The method replaces the need to select and deploy a fixed-size model variant (S/B/L) by converting one pre-trained transformer into an elastic model that serves a continuum of sparsity levels, using a differentiable relaxation that jointly ranks every MLP neuron and attention head on a common scale. It costs a short construction process (reported as minutes) plus validation per sparsity target; no retraining or weight correction is required, and storage of the original dense weights is preserved with structured units removed at serving time. Failure modes include steeper accuracy loss beyond the reported 60% sparsity regime, uneven behavior on tasks or model families outside the 27 evaluated encoders, and hardware-specific speedups that may not reproduce on the reader's GPU without structured-pruning-aware kernels.", "rationale": "It is directly usable with a single GPU and offers measurable latency and throughput gains, but the reader should validate accuracy degradation at their chosen sparsity level on their own tasks before production adoption.", "technique": "Retraining-free elastic structured pruning via joint unit ranking", "family": "esparsidade_e_poda", "recommendation": "testar", "published": "2026-10-14", "as_of": "2026-10-18", "editorial_status": "indexed", "source_url": "https://arxiv.org/abs/2610.20588", "independent_implementations": 0, "total_implementations": 0, "stars": 0, "citations": null, "claims": [ { "claim_id": "claim-01", "statement": "The method replaces the need to select and deploy a fixed-size model variant (S/B/L) by converting one pre-trained transformer into an elastic model that serves a continuum of sparsity levels, using a differentiable relaxation that jointly ranks every MLP neuron and attention head on a common scale. It costs a short construction process (reported as minutes) plus validation per sparsity target; no retraining or weight correction is required, and storage of the original dense weights is preserved with structured units removed at serving time. Failure modes include steeper accuracy loss beyond the reported 60% sparsity regime, uneven behavior on tasks or model families outside the 27 evaluated encoders, and hardware-specific speedups that may not reproduce on the reader's GPU without structured-pruning-aware kernels.", "basis": "inferred", "result": "", "baseline": "", "conditions": "", "source_url": "https://arxiv.org/abs/2610.20588", "source_page": null, "source_excerpt": "" }, { "claim_id": "claim-02", "statement": "At 60% sparsity, a pruned AugReg ViT-B/16 achieves up to 1.80x higher throughput and 1.26x lower single-query latency on an A100 GPU; DINOv3 ViT-H+/16 pruned to 50% sparsity loses only 0.8% ImageNet-1k linear-probing accuracy.", "basis": "inferred", "result": "", "baseline": "", "conditions": "", "source_url": "https://arxiv.org/abs/2610.20588", "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": "" }