{ "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": "2609.35139", "title": "CacheRepair: Learning to Repair Cross-Chunk Context in RAG for KV Cache Fusion", "summary": "CacheRepair replaces full joint prefill over retrieved chunks (and expensive selective recomputation) with independent per-chunk KV precomputation plus a small learned network that predicts residuals restoring cross-chunk context. The cost is a per-target-LLM training run of the repair network on a retrieval corpus, online cache transfer plus repair latency already included in TTFT, and added serving complexity. It can fail when the frozen target LLM or retrieval distribution changes (requiring retraining), and quality gains were not uniform across all model-dataset pairs, so small repairers may recover less cross-chunk context than reported best cases.", "rationale": "The method is deployable on a single GPU with a frozen LLM and a small trained repairer, but its quality-latency benefit depends on the reader's specific model and retrieval workload, so a workload-specific pilot is warranted before adoption.", "technique": "Learned repair of independently precomputed RAG KV caches", "family": "cache_kv", "recommendation": "testar", "published": "2026-09-28", "as_of": "2026-10-18", "editorial_status": "indexed", "source_url": "https://arxiv.org/abs/2609.35139", "independent_implementations": 0, "total_implementations": 1, "stars": 1, "citations": null, "claims": [ { "claim_id": "claim-01", "statement": "CacheRepair replaces full joint prefill over retrieved chunks (and expensive selective recomputation) with independent per-chunk KV precomputation plus a small learned network that predicts residuals restoring cross-chunk context. The cost is a per-target-LLM training run of the repair network on a retrieval corpus, online cache transfer plus repair latency already included in TTFT, and added serving complexity. It can fail when the frozen target LLM or retrieval distribution changes (requiring retraining), and quality gains were not uniform across all model-dataset pairs, so small repairers may recover less cross-chunk context than reported best cases.", "basis": "inferred", "result": "", "baseline": "", "conditions": "", "source_url": "https://arxiv.org/abs/2609.35139", "source_page": null, "source_excerpt": "" }, { "claim_id": "claim-02", "statement": "Largest repairers achieve 1.69-4.61x speedups in median TTFT over full prefill and improve mean F1 by 2.1-26.1 percentage points over direct cache reuse, across three LLMs and four datasets.", "basis": "inferred", "result": "", "baseline": "", "conditions": "", "source_url": "https://arxiv.org/abs/2609.35139", "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": "" }