{
  "schema_version": 1,
  "builder_plan": {
    "scenario": "An agent looks good in demos, but you cannot tell which product tasks it fails.",
    "change": "Use the study's evaluation idea to inspect a small set of tasks with human-checked answers.",
    "baseline": "Your current evaluator and a manually reviewed reference set.",
    "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": "Agreement with human review, missed failures, false alarms, and evaluation cost per 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": "2610.30047",
  "title": "SWE-DIVE: Multi-Turn Coding Agent Evaluation with Dynamic User Simulation",
  "summary": "SWE-DIVE replaces static, one-shot GitHub-issue benchmarks (e.g., SWE-bench) with dynamic, multi-turn user simulation to evaluate coding agents. It introduces complexity by requiring a user simulator, conversation validator, and orchestration environment (Harbor), which increases setup overhead and evaluation cost compared to single-turn benchmarks. Potential failure modes include unrealistic user behavior simulation, validator misalignment with true task requirements, and high infrastructure demands for continuous orchestration that may exceed constrained budgets.",
  "rationale": "The framework is designed for academic benchmarking and requires multi-component orchestration that may exceed constrained infrastructure budgets; while relevant for agent evaluation, its adoption requires validation of simulator fidelity and cost-effectiveness at small scale.",
  "technique": "SWE-DIVE: Multi-Turn Coding Agent Evaluation Framework",
  "family": "avaliacao_de_agente",
  "recommendation": "observar",
  "published": "2026-10-21",
  "as_of": "2026-10-25",
  "editorial_status": "indexed",
  "source_url": "https://arxiv.org/abs/2610.30047",
  "independent_implementations": 0,
  "total_implementations": 1,
  "stars": 0,
  "citations": 0,
  "claims": [
    {
      "claim_id": "claim-01",
      "statement": "SWE-DIVE replaces static, one-shot GitHub-issue benchmarks (e.g., SWE-bench) with dynamic, multi-turn user simulation to evaluate coding agents. It introduces complexity by requiring a user simulator, conversation validator, and orchestration environment (Harbor), which increases setup overhead and evaluation cost compared to single-turn benchmarks. Potential failure modes include unrealistic user behavior simulation, validator misalignment with true task requirements, and high infrastructure demands for continuous orchestration that may exceed constrained budgets.",
      "basis": "inferred",
      "result": "",
      "baseline": "",
      "conditions": "",
      "source_url": "https://arxiv.org/abs/2610.30047",
      "source_page": null,
      "source_excerpt": ""
    },
    {
      "claim_id": "claim-02",
      "statement": "No quantitative performance claim is provided in the abstract; the framework proposes a qualitative methodology for dynamic, multi-turn evaluation.",
      "basis": "inferred",
      "result": "",
      "baseline": "",
      "conditions": "",
      "source_url": "https://arxiv.org/abs/2610.30047",
      "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": ""
}