{ "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": "2610.11602", "title": "Where Do the Tokens Go? Understanding and Reducing Costs in LLM Agents for Vulnerability Discovery", "summary": "AVRI replaces the agent's repeated ad hoc retrieval and reconstruction of code evidence with a persistent Bidirectional Evidence Trace (BET) that links harness input consumption to unsafe-operation conditions, exposed through reading, analysis, and persistence commands. It costs the engineering effort of building and maintaining a task-specific agent interface and structured trace format, adds auxiliary command overhead per step, and its gains are demonstrated only on 20 CyberGym tasks with two agents. It can fail if the trace retains incorrect or stale hypotheses that misdirect vulnerability reasoning, or if the interface design does not transfer to other agent frameworks or vulnerability classes.", "rationale": "The cost reduction is real but modest, validated on a small task set in a niche security-agent domain, so a practitioner should pilot it on their own agentic workloads before committing to the interface-building effort.", "technique": "Persistent bidirectional evidence trace for security agents", "family": "memoria_e_contexto", "recommendation": "testar", "published": "2026-10-08", "as_of": "2026-10-18", "editorial_status": "indexed", "source_url": "https://arxiv.org/abs/2610.11602", "independent_implementations": 0, "total_implementations": 0, "stars": 0, "citations": 0, "claims": [ { "claim_id": "claim-01", "statement": "AVRI replaces the agent's repeated ad hoc retrieval and reconstruction of code evidence with a persistent Bidirectional Evidence Trace (BET) that links harness input consumption to unsafe-operation conditions, exposed through reading, analysis, and persistence commands. It costs the engineering effort of building and maintaining a task-specific agent interface and structured trace format, adds auxiliary command overhead per step, and its gains are demonstrated only on 20 CyberGym tasks with two agents. It can fail if the trace retains incorrect or stale hypotheses that misdirect vulnerability reasoning, or if the interface design does not transfer to other agent frameworks or vulnerability classes.", "basis": "inferred", "result": "", "baseline": "", "conditions": "", "source_url": "https://arxiv.org/abs/2610.11602", "source_page": null, "source_excerpt": "" }, { "claim_id": "claim-02", "statement": "On 20 tasks, AVRI reduces total token cost by 18.0% for Codex and 23.7% for OpenCode while preserving success rates and improving or maintaining recall.", "basis": "inferred", "result": "", "baseline": "", "conditions": "", "source_url": "https://arxiv.org/abs/2610.11602", "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": "" }