{ "schema_version": 1, "builder_plan": { "scenario": "An assistant chooses the wrong API or sends arguments your product cannot use.", "change": "Apply the proposed tool-selection or argument-generation change in a sandbox.", "baseline": "Your current agent with the same tool schemas and task budget.", "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": "Correct completed actions, invalid calls, retries, latency, and cost per completed task.", "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": "2610.28282", "title": "ChartEvo: Self-Evolving Agent Harnesses for Long-Tail Chart Reasoning", "summary": "It replaces manually engineered per-chart-type strategies and tools with a recursive framework that evolves reusable skills, executable tools, and tool-call chains driven by observed failures. It costs additional inference overhead from tool-augmented reasoning (partially controlled via learned local tool-call chains), plus the engineering complexity of maintaining an evolving skill and tool library around an API-based MLLM. It can fail if the self-improvement loop overfits to the evaluation distribution, if evolved tools are unreliable or unsafe to execute, and the reported gains are validated only on two chart benchmarks with a single proprietary base model.", "rationale": "The approach is feasible with third-party APIs and no training, but the self-evolution machinery adds operational complexity and its gains need validation on the reader's actual chart distribution before production adoption.", "technique": "Self-evolving tool and skill harness for chart agents", "family": "uso_de_ferramenta", "recommendation": "testar", "published": "2026-10-20", "as_of": "2026-10-21", "editorial_status": "indexed", "source_url": "https://arxiv.org/abs/2610.28282", "independent_implementations": 1, "total_implementations": 2, "stars": 132, "citations": null, "claims": [ { "claim_id": "claim-01", "statement": "It replaces manually engineered per-chart-type strategies and tools with a recursive framework that evolves reusable skills, executable tools, and tool-call chains driven by observed failures. It costs additional inference overhead from tool-augmented reasoning (partially controlled via learned local tool-call chains), plus the engineering complexity of maintaining an evolving skill and tool library around an API-based MLLM. It can fail if the self-improvement loop overfits to the evaluation distribution, if evolved tools are unreliable or unsafe to execute, and the reported gains are validated only on two chart benchmarks with a single proprietary base model.", "basis": "inferred", "result": "", "baseline": "", "conditions": "", "source_url": "https://arxiv.org/abs/2610.28282", "source_page": null, "source_excerpt": "" }, { "claim_id": "claim-02", "statement": "On long-tail charts from ChartBench and ChartX, ChartEvo improves accuracy over its base MLLM, Qwen3.7-Plus, by 45.6% and 32.5%, respectively.", "basis": "inferred", "result": "", "baseline": "", "conditions": "", "source_url": "https://arxiv.org/abs/2610.28282", "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": "" }