{ "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": "2609.17107", "title": "Symbolic Separation: Grounding Deep Agents in Knowledge Graphs for Trustworthy Operational Data Analytics", "summary": "The method replaces LLM-inferred joins and free-form tool calls over heterogeneous data sources with an ontology-constrained Virtual Knowledge Graph and deterministic pre-execution validation, so a question becomes one validated graph traversal. It costs the upfront effort of building and maintaining a domain ontology and mapping data sources into the graph, plus added system complexity around the validation layer. It can fail when the ontology is incomplete or drifts from the underlying schema, when questions fall outside the modeled domain, or when the validation layer rejects legitimate queries, and the reported gains come from a single telemetry domain and may not transfer.", "rationale": "The approach fits constrained infrastructure since it lets a smaller on-premise model match a larger one at lower token cost, but it requires workload-specific ontology engineering whose payoff must be validated on the reader's own data sources.", "technique": "Ontology-constrained Virtual Knowledge Graph for analytics agents", "family": "uso_de_ferramenta", "recommendation": "testar", "published": "2026-09-15", "as_of": "2026-10-18", "editorial_status": "indexed", "source_url": "https://arxiv.org/abs/2609.17107", "independent_implementations": 0, "total_implementations": 0, "stars": 0, "citations": 0, "claims": [ { "claim_id": "claim-01", "statement": "The method replaces LLM-inferred joins and free-form tool calls over heterogeneous data sources with an ontology-constrained Virtual Knowledge Graph and deterministic pre-execution validation, so a question becomes one validated graph traversal. It costs the upfront effort of building and maintaining a domain ontology and mapping data sources into the graph, plus added system complexity around the validation layer. It can fail when the ontology is incomplete or drifts from the underlying schema, when questions fall outside the modeled domain, or when the validation layer rejects legitimate queries, and the reported gains come from a single telemetry domain and may not transfer.", "basis": "inferred", "result": "", "baseline": "", "conditions": "", "source_url": "https://arxiv.org/abs/2609.17107", "source_page": null, "source_excerpt": "" }, { "claim_id": "claim-02", "statement": "Raises end-to-end task success from 43% to 86% versus a non-symbolic ablation and cuts token cost by 2.4x on 49.9 TB of telemetry, enabling a smaller on-premise model to outperform a larger one.", "basis": "inferred", "result": "", "baseline": "", "conditions": "", "source_url": "https://arxiv.org/abs/2609.17107", "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": [ { "anchor": "S3.Ex1", "label": "", "section": "III Methodology", "role": "proposed_method", "latex": "\\delta(s_{i}^{\\ast})=\\begin{cases}\\texttt{fail},&s_{i}^{\\ast}<0.5,\\\\[2.0pt]\n\\texttt{clarification},&0.5\\leq s_{i}^{\\ast}<0.7,\\\\[2.0pt]\n\\texttt{accept},&s_{i}^{\\ast}\\geq 0.7.\\end{cases}", "mathml": "δ⁡(si∗)={fail,si∗<0.5,clarification,0.5≤si∗<0.7,accept,si∗≥0.7.", "context": "A decision function δ⁡(si∗) determines the retrieval outcome:", "headline": "", "explanation": "", "objective": "", "symbols": {}, "evidence": "", "evidence_url": "", "source_url": "" } ], "equations_status": "selected", "core_kind": "formula", "equations_fetched_at": "2026-10-17", "equations_source_url": "", "equations_source_sha256": "", "exposures_reviewed_at": "", "exposures_source_sha256": "" }