{ "schema_version": 1, "builder_plan": { "scenario": "Long customer conversations exhaust memory or limit concurrent requests.", "change": "Change only the attention cache policy, after checking model and runtime support.", "baseline": "The same model with its ordinary attention cache.", "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": "Peak cache memory, long-context answer quality, first-token latency, and requests completed.", "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": "2610.09827", "title": "Dual-QK: Sharp Queries and Flat Keys for Prunable 2-bit KV Caches", "summary": "Dual-QK replaces standard rotation-based INT2 KV-cache quantization (e.g., OSCAR-style orthogonal rotations) with paired non-orthogonal query and key transforms that whiten keys for 2-bit storage while concentrating query energy so channels can be pruned dynamically. It costs a calibration pass over query and key statistics, a modified attention implementation (the paper uses SGLang), and the integration complexity of channel-0 protection and bucket-relative RoPE, with accuracy losses possible on tasks sensitive to 40% query-channel pruning. It can fail if the workload's context lengths or query statistics diverge from calibration data, if the serving stack cannot accommodate the custom kernel, or if INT2 plus pruning degrades accuracy on the reader's specific models and benchmarks, which the paper covers only partially.", "rationale": "The memory and throughput gains directly address constrained-infrastructure inference, but the method requires calibration, a custom serving implementation, and workload-specific accuracy validation before production adoption.", "technique": "Dual-QK: paired transforms for 2-bit prunable KV cache", "family": "cache_kv", "recommendation": "testar", "published": "2026-10-07", "as_of": "2026-10-18", "editorial_status": "indexed", "source_url": "https://arxiv.org/abs/2610.09827", "independent_implementations": 0, "total_implementations": 0, "stars": 0, "citations": null, "claims": [ { "claim_id": "claim-01", "statement": "Dual-QK replaces standard rotation-based INT2 KV-cache quantization (e.g., OSCAR-style orthogonal rotations) with paired non-orthogonal query and key transforms that whiten keys for 2-bit storage while concentrating query energy so channels can be pruned dynamically. It costs a calibration pass over query and key statistics, a modified attention implementation (the paper uses SGLang), and the integration complexity of channel-0 protection and bucket-relative RoPE, with accuracy losses possible on tasks sensitive to 40% query-channel pruning. It can fail if the workload's context lengths or query statistics diverge from calibration data, if the serving stack cannot accommodate the custom kernel, or if INT2 plus pruning degrades accuracy on the reader's specific models and benchmarks, which the paper covers only partially.", "basis": "inferred", "result": "", "baseline": "", "conditions": "", "source_url": "https://arxiv.org/abs/2610.09827", "source_page": null, "source_excerpt": "" }, { "claim_id": "claim-02", "statement": "At 128K context, 6.8x KV-cache compression and an estimated 8.3x reduction in KV read volume versus unpruned BF16, with up to 3.75x decoding throughput in the authors' SGLang implementation.", "basis": "inferred", "result": "", "baseline": "", "conditions": "", "source_url": "https://arxiv.org/abs/2610.09827", "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": "" }