{ "schema_version": 1, "builder_plan": { "scenario": "A support or data assistant repeats work across related customer requests.", "change": "Try the paper's memory change in a replay of past tasks, with separate memory per customer.", "baseline": "Your current memory pipeline plus a no-memory control using the same agent.", "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 success, stale memories, retrieval misses, response latency, and all memory maintenance costs.", "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": "2610.26836", "title": "Same Simulation, Different Evidence: Where Retrieved Tool Output Costs LLM Agents Accuracy", "summary": "The study evaluates replacing in-context tool outputs with file-based retrieval (section indexes) in agent harnesses, and proposes mitigations: explicitly naming the authoritative output file or inlining outputs that fit. Retrieval saves context window space, but for GPT-5.x models it costs 8-11 accuracy points; the naming prompt costs essentially nothing and recovers most of the loss. The residual failure mode is evidence substitution: the agent computes correctly but answers from its own higher-precision intermediate files, misrounding to a neighboring option, and prompting for correct rounding does not fix this.", "rationale": "The reader can apply the finding immediately in API-based agent harnesses at no infrastructure cost: pin the authoritative tool output in the prompt or inline outputs that fit, rather than relying on undirected retrieval over saved files.", "technique": "Authoritative-output naming in agent context management", "family": "memoria_e_contexto", "recommendation": "adotar", "published": "2026-10-19", "as_of": "2026-10-20", "editorial_status": "indexed", "source_url": "https://arxiv.org/abs/2610.26836", "independent_implementations": 2, "total_implementations": 3, "stars": 132, "citations": null, "claims": [ { "claim_id": "claim-01", "statement": "The study evaluates replacing in-context tool outputs with file-based retrieval (section indexes) in agent harnesses, and proposes mitigations: explicitly naming the authoritative output file or inlining outputs that fit. Retrieval saves context window space, but for GPT-5.x models it costs 8-11 accuracy points; the naming prompt costs essentially nothing and recovers most of the loss. The residual failure mode is evidence substitution: the agent computes correctly but answers from its own higher-precision intermediate files, misrounding to a neighboring option, and prompting for correct rounding does not fix this.", "basis": "inferred", "result": "", "baseline": "", "conditions": "", "source_url": "https://arxiv.org/abs/2610.26836", "source_page": null, "source_excerpt": "" }, { "claim_id": "claim-02", "statement": "Reading simulator output via a section index instead of in-context costs GPT-5.1 11.3 and GPT-5.2 7.9 accuracy points (Claude models unaffected); naming the authoritative output file recovers 6.3-7.9 points over plain retrieval, and inlining outputs that fit matches this for GPT-5.1/5.2.", "basis": "inferred", "result": "", "baseline": "", "conditions": "", "source_url": "https://arxiv.org/abs/2610.26836", "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": "" }