{"record": {"aliases": [], "artifacts": [{"id": "arxiv:2609.12623", "kind": "paper", "url": "https://arxiv.org/abs/2609.12623"}], "authors": ["Utkarsh Tyagi", "Ramaneswaran Selvakumar", "Advait Gosai", "Sonal Kumar", "Nikhil Barhate", "Isabell Sagar", "Steven Li", "Miheer Bavare", "Daniel Quigley", "Fabiola Tapia Carrillo", "Jose M Patron E", "Diego Macías Gutiérrez", "Paul Song", "Ramani Duraiswami", "Dinesh Manocha", "Yunzhong He"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2609.12623", "source_url": "https://arxiv.org/abs/2609.12623"}, "categories": ["General AI", "Speech & Audio", "cs.AI"], "description": {"en": "SteerBench evaluates full-duplex spoken steerability using 390 spoken prompts and 1,067 human-authored binary audio and text rubrics spanning tone, persona, style or accent, and speed or length."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2609.12623", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2609.12623", "title": "SteerBench"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "institutions": ["Hochschule Aalen", "University of Maryland, College Park"], "institutions_reference": {"basis": "openalex_linked_institutions", "doi": "10.48550/arxiv.2609.12623", "retrieved_at": "2026-11-04", "source_key": "claire-radar:2609.12623", "source_url": "https://openalex.org/W7212887793"}, "key": "claire-radar:2609.12623", "modality": null, "name": "SteerBench", "openness": {"code_license": null, "data_license": null, "evidence": [], "status": "unknown"}, "provenance": {"confidence": "High", "crawl_bundle": "claire_radar_2026-10-14", "crawled_at": "2026-10-14T09:44:51Z", "data_status": "primary-source-indexed", "display_eligible": "false", "origin_record_id": "bm_steerbench_3e1ad022", "origin_source": "arxiv", "recognition_confidence": "1.0", "record_sha256": "e8f92b75800bfd5d2422512fd2f2bbea74348a15b46673d9337ba587a8d0ede4", "relation": "introduces", "review_model": "deepseek-v4-pro", "review_state": "ai-name-audit-deferred", "reviewed_at": "2026-09-15T09:35:32.523429Z", "source_url": "https://arxiv.org/abs/2609.12623"}, "publisher": null, "released": "2026-09-11", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2609.12623", "source_url": "https://arxiv.org/abs/2609.12623"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2609-12623", "source": "claire_radar", "source_benchmark_id": "2609.12623", "source_metadata": {"claire_radar": {"adoption30d": null, "aliases": [], "annotation": "Machine Generated", "applicationDomains": ["General AI"], "area": "Speech & Audio", "attention": {"asOf": "2026-10-05", "attemptedAt": "2026-10-04T15:31:48.030916Z", "githubRepo": null, "githubScope": null, "githubStars": null, "hfDailySubmittedAt": "2026-09-21T00:00:00.000Z", "hfDataset": null, "hfDatasetDownloads": null, "hfDatasetLikes": null, "hfPaperUpvotes": 6, "hfPaperUrl": "https://huggingface.co/papers/2609.12623", "observedAt": "2026-10-04T15:31:48.030916Z", "signalStatus": {"githubStars": {"state": "not_applicable"}, "hfDatasetDownloads": {"state": "not_applicable"}, "hfDatasetLikes": {"state": "not_applicable"}, "hfPaperUpvotes": {"observedAt": "2026-10-04T15:31:48.030916Z", "state": "fresh"}}, "status": "fresh"}, "attentionForecast": {"confidence": "Low", "horizon": "7d", "reason": "Only an in-paper benchmark description is supplied with no released artifacts or public path, limiting expected first-week attention.", "score": 0}, "capabilities": [], "confidence": "High", "construction": "Original Synthetic", "constructionDetail": "Unknown — the indexer does not infer construction details without explicit source evidence.", "copyGeneration": {"generatedAt": "2026-09-15T09:37:10.086794Z", "inputHash": "daadb38ed75a7fc2ef93ad511f3d5907c77c309a334976e02ace5a1cd9e0a66f", "model": "deepseek-v4-flash"}, "curation": {"decisionReason": "The abstract mentions the benchmark but no code, data, or public submission path is provided, so the evaluation and scoring contract cannot be confirmed.", "model": "deepseek-v4-pro", "policy": "source-evidence-v2", "reviewedAt": "2026-09-15T09:35:32.523429Z", "state": "ai-name-audit-deferred"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "SteerBench evaluates full-duplex spoken steerability using 390 spoken prompts and 1,067 human-authored binary audio and text rubrics spanning tone, persona, style or accent, and speed or length.", "displayEligible": false, "domainCuration": {"method": "rules-v1", "state": "auto"}, "evaluationMode": "public_reusable", "evidence": {"reasonCodes": ["exact named benchmark artifact released in abstract", "benchmark release stated in one sentence", "evaluation protocol evidence"], "snippet": "To evaluate full-duplex spoken steerability, we introduce SteerBench, a benchmark with 390 spoken prompts and 1,067 human-authored binary audio and text rubrics spanning tone, persona, style/accent, and speed/length."}, "familyId": "bmf_5016ef219bef", "firstSeenAt": "2026-09-14", "heat": null, "id": "bm_steerbench_3e1ad022", "indexedAt": "2026-09-14T00:39:56.575498Z", "industrySectors": [], "links": {"code": null, "data": null, "paper": "https://arxiv.org/abs/2609.12623", "pdf": "https://arxiv.org/pdf/2609.12623", "project": null, "report": "https://arxiv.org/abs/2609.12623"}, "metrics": [], "motivation": "Full-duplex spoken dialogue models support low-latency turn taking, interruption handling, and backchanneling, yet a key capability remains underexplored: steerability, the ability to reliably shift conversational behavior along attributes such as tone, persona, speaking rate, and voice style in response to user instructions.", "name": "SteerBench", "oneLine": "SteerBench evaluates full-duplex spoken steerability using 390 spoken prompts and 1,067 human-authored binary audio and text rubrics spanning tone, persona, style or accent, and speed or length.", "paperTitle": "SteerDuplex: Steerable Duplex Speech Dialogue Models", "primaryDomain": "General AI", "publication": {"evidence": null, "evidenceLevel": "unverified", "evidenceUrl": "https://arxiv.org/abs/2609.12623", "source": "arxiv-metadata", "status": "unverified", "venue": null, "verifiedAt": "2026-09-14T00:39:56.575498Z"}, "publications": [], "ranking": {"30d": {"confidence": "Low", "coverage": 0.3, "growth": {"components": {"githubStars": {"normalization": "percentile", "normalized": null, "percentile": null, "value": null}, "hfDatasetDownloads": {"normalization": "percentile", "normalized": null, "percentile": null, "value": null}, "hfPaperUpvotes": {"normalization": "percentile", "normalized": null, "percentile": null, "value": null}}, "confidence": "Low", "coverage": 0.0, "elapsedDays": 30, "historyDate": "2026-09-05", "rank": null, "score": null, "windowDays": 30}, "level": {"components": {"githubStars": {"normalization": "log-max", "normalized": null, "value": null}, "hfDatasetDownloads": {"normalization": "log-max", "normalized": null, "value": null}, "hfPaperUpvotes": {"normalization": "log-max", "normalized": 0.36232383170565413, "value": 6}}, "confidence": "Low", "coverage": 0.3, "rank": null, "score": 46}, "method": "attention-ranking-v13", "rank": null, "score": 46}, "90d": {"confidence": "Low", "coverage": 0.15, "growth": {"components": {"githubStars": {"normalization": "percentile", "normalized": null, "percentile": null, "value": null}, "hfDatasetDownloads": {"normalization": "percentile", "normalized": null, "percentile": null, "value": null}, "hfPaperUpvotes": {"normalization": "percentile", "normalized": null, "percentile": null, "value": null}}, "confidence": "Low", "coverage": 0.0, "elapsedDays": null, "historyDate": null, "rank": null, "score": null, "windowDays": 90}, "level": {"components": {"githubStars": {"normalization": "log-max", "normalized": null, "value": null}, "hfDatasetDownloads": {"normalization": "log-max", "normalized": null, "value": null}, "hfPaperUpvotes": {"normalization": "log-max", "normalized": 0.34383183387990046, "value": 6}}, "confidence": "Low", "coverage": 0.15, "rank": null, "score": 48}, "method": "attention-ranking-v13", "rank": null, "score": 48}}, "readiness": "Paper only", "recognitionConfidence": 1.0, "relation": "introduces", "releasedAt": "2026-09-11", "source": {"authors": ["Utkarsh Tyagi", "Ramaneswaran Selvakumar", "Advait Gosai", "Sonal Kumar", "Nikhil Barhate", "Isabell Sagar", "Steven Li", "Miheer Bavare", "Daniel Quigley", "Fabiola Tapia Carrillo", "Jose M Patron E", "Diego Mac\\'ias Guti\\'errez", "Paul Song", "Ramani Duraiswami", "Dinesh Manocha", "Yunzhong He"], "categories": ["cs.AI", "cs.CL"], "id": "2609.12623", "title": "SteerDuplex: Steerable Duplex Speech Dialogue Models", "type": "arxiv", "url": "https://arxiv.org/abs/2609.12623"}, "sourceUpdatedAt": "2026-09-11T09:22:46Z", "topics": ["cs.AI"], "venueAttempts": [], "whyItMatters": "Full-duplex speech systems require evaluation of controllable spoken attributes rather than text-only quality; a rubric-based benchmark could support comparison of steerability across models."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
