{ "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": "2609.40198", "title": "SCB: SpeechConversationBench for Evaluating Multi-Turn Reasoning in Speech-to-Speech Models", "summary": "SCB replaces ad hoc or single-turn evaluation of speech-to-speech systems with a controlled three-condition protocol (full, concat, sharded) that isolates degradation caused by incremental spoken disclosure on 103 GSM8K-derived problems. It costs only evaluation engineering effort: shard construction, multi-turn harness implementation, and API spend across at least two runs per condition, with no model training or additional GPU requirements. It can fail as a selection tool because 103 math problems cover a narrow domain, results may not transfer to non-mathematical voice workloads, and the top-scoring system is a proprietary pipeline whose context-management design is not available for replication.", "rationale": "The reader deploying speech-to-speech systems via third-party APIs can reuse the sharded-versus-concat evaluation pattern cheaply to detect multi-turn degradation, but the benchmark itself is narrow and its best result depends on an unreleased proprietary system, so it warrants workload-specific validation rather than direct adoption.", "technique": "Sharded multi-turn spoken reasoning benchmark (SCB)", "family": "avaliacao_de_agente", "recommendation": "testar", "published": "2026-09-30", "as_of": "2026-10-18", "editorial_status": "indexed", "source_url": "https://arxiv.org/abs/2609.40198", "independent_implementations": 0, "total_implementations": 1, "stars": 1, "citations": null, "claims": [ { "claim_id": "claim-01", "statement": "SCB replaces ad hoc or single-turn evaluation of speech-to-speech systems with a controlled three-condition protocol (full, concat, sharded) that isolates degradation caused by incremental spoken disclosure on 103 GSM8K-derived problems. It costs only evaluation engineering effort: shard construction, multi-turn harness implementation, and API spend across at least two runs per condition, with no model training or additional GPU requirements. It can fail as a selection tool because 103 math problems cover a narrow domain, results may not transfer to non-mathematical voice workloads, and the top-scoring system is a proprietary pipeline whose context-management design is not available for replication.", "basis": "inferred", "result": "", "baseline": "", "conditions": "", "source_url": "https://arxiv.org/abs/2609.40198", "source_page": null, "source_excerpt": "" }, { "claim_id": "claim-02", "statement": "Sharded multi-turn delivery reduces final-answer accuracy by 5.0-25.3 percentage points relative to concatenated single-turn delivery across four commercial speech systems; the proprietary LEGO pipeline with explicit context management holds 77.5% accuracy across all conditions versus 76.6% sharded for GPT-4o Realtime.", "basis": "inferred", "result": "", "baseline": "", "conditions": "", "source_url": "https://arxiv.org/abs/2609.40198", "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": "" }