{"record": {"aliases": [], "artifacts": [{"id": "arxiv:2605.28035", "kind": "paper", "url": "https://arxiv.org/abs/2605.28035"}], "authors": ["Haitian Li", "Yanghao Zhou", "Heyan Huang", "Liangji Chen", "YiMing Cheng", "Xu Liu", "Dian Jin", "Jiajun Xu", "Jingyun Liao", "Tian Lan", "Ziqin Zhou", "Yueying Liu", "Yu Bai", "Changsen Yuan", "Jinxing Zhou", "Xian-Ling Mao", "Xuefeng Chen", "Yousheng Feng"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2605.28035", "source_url": "https://arxiv.org/abs/2605.28035"}, "categories": ["General AI", "Vision & 3D", "cs.AI"], "description": {"en": "MTAVG-Bench 2.0 evaluates omni large language models on diagnosing high-level cinematic failures in multi-talker audio-video generation, with over 10,000 QA instances covering acting, narrative, atmosphere, and audio-visual language."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2605.28035", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2605.28035", "title": "MTAVG-Bench"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "key": "claire-radar:2605.28035", "modality": null, "name": "MTAVG-Bench", "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": "unknown", "origin_record_id": "bm_mtavg-bench_0372b3a9", "origin_source": "arxiv", "recognition_confidence": "1.0", "record_sha256": "00d2df29181a3ba6a7f49620ec142f9337f3085af9582abf4a30ddbe41b93352", "relation": "introduces", "review_state": "unreviewed", "source_url": "https://arxiv.org/abs/2605.28035"}, "publisher": null, "released": "2026-05-27", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2605.28035", "source_url": "https://arxiv.org/abs/2605.28035"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2605-28035", "source": "claire_radar", "source_benchmark_id": "2605.28035", "source_metadata": {"claire_radar": {"adoption30d": null, "aliases": [], "annotation": "Unknown", "applicationDomains": ["General AI"], "area": "Vision & 3D", "attention": {"asOf": "2026-10-05", "attemptedAt": "2026-10-04T15:31:48.030916Z", "githubRepo": null, "githubScope": null, "githubStars": null, "hfDailySubmittedAt": null, "hfDataset": null, "hfDatasetDownloads": null, "hfDatasetLikes": null, "hfPaperUpvotes": null, "hfPaperUrl": "https://huggingface.co/papers/2605.28035", "observedAt": "2026-10-04T03:02:42.754940Z", "signalStatus": {"githubStars": {"state": "not_applicable"}, "hfDatasetDownloads": {"state": "not_applicable"}, "hfDatasetLikes": {"state": "not_applicable"}, "hfPaperUpvotes": {"checkedAt": "2026-10-04T15:31:48.030916Z", "state": "unavailable"}}, "status": "unavailable"}, "capabilities": [], "confidence": "High", "construction": "Unknown", "constructionDetail": "Unknown — the indexer does not infer construction details without explicit source evidence.", "copyGeneration": {"generatedAt": "2026-08-20T15:22:10.044731Z", "inputHash": "0dbaa6e642f474af1f621756a2d6f20f876f9c4ea1b54424b218d5de7b88e490", "model": "deepseek-v4-flash"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "MTAVG-Bench 2.0 evaluates omni large language models on diagnosing high-level cinematic failures in multi-talker audio-video generation, with over 10,000 QA instances covering acting, narrative, atmosphere, and audio-visual language.", "domainCuration": {"method": "rules-v1", "state": "auto"}, "evidence": {"reasonCodes": ["exact coined title identity tied to benchmark evidence", "benchmark release stated in one sentence", "evaluation protocol evidence"], "snippet": "To fill this gap, we introduce MTAVG-Bench 2.0, a benchmark for diagnosing failure modes of cinematic expressiveness in multi-talker audio-video generation."}, "familyId": "bmf_ac59d735d593", "firstSeenAt": "2026-08-19", "githubIndex": {"candidateCount": 0, "indexedAt": "2026-09-17T04:40:34.111665+00:00", "repositories": [], "status": "not-found"}, "heat": null, "id": "bm_mtavg-bench_0372b3a9", "indexedAt": "2026-08-19T11:05:13.395391Z", "industrySectors": [], "links": {"code": null, "data": null, "hfPaper": "https://huggingface.co/papers/2605.28035", "paper": "https://arxiv.org/abs/2605.28035", "pdf": "https://arxiv.org/pdf/2605.28035", "project": null, "report": "https://arxiv.org/abs/2605.28035"}, "metrics": [], "motivation": "In recent years, Multi-Talker Audio-Video Generation (MTAVG) models have shown promising performance on fundamental metrics such as lip-sync and audio-visual alignment.", "name": "MTAVG-Bench", "oneLine": "MTAVG-Bench 2.0 evaluates omni large language models on diagnosing high-level cinematic failures in multi-talker audio-video generation, with over 10,000 QA instances covering acting, narrative, atmosphere, and audio-visual language.", "paperTitle": "MTAVG-Bench 2.0: Diagnosing Failure Modes of Cinematic Expressiveness in Multi-Talker Audio-Video Generation", "primaryDomain": "General AI", "publication": {"evidence": null, "evidenceLevel": "unverified", "evidenceUrl": "https://arxiv.org/abs/2605.28035", "source": "arxiv-metadata", "status": "unverified", "venue": null, "verifiedAt": "2026-10-14T01:46:58.058199Z"}, "publications": [], "ranking": {}, "readiness": "Paper only", "recognitionConfidence": 1.0, "relation": "introduces", "releasedAt": "2026-05-27", "source": {"authors": ["Haitian Li", "Yanghao Zhou", "Heyan Huang", "Liangji Chen", "YiMing Cheng", "Xu Liu", "Dian Jin", "Jiajun Xu", "Jingyun Liao", "Tian Lan", "Ziqin Zhou", "Yueying Liu", "Yu Bai", "Changsen Yuan", "Jinxing Zhou", "Xian-Ling Mao", "Xuefeng Chen", "Yousheng Feng"], "categories": ["cs.AI", "cs.MM", "cs.SD"], "id": "2605.28035", "title": "MTAVG-Bench 2.0: Diagnosing Failure Modes of Cinematic Expressiveness in Multi-Talker Audio-Video Generation", "type": "arxiv", "url": "https://arxiv.org/abs/2605.28035"}, "sourceUpdatedAt": "2026-05-27T06:38:54Z", "topics": ["cs.AI"], "venueAttempts": [], "whyItMatters": "Standard metrics like lip-sync do not capture cinematic expressiveness; this benchmark targets a gap in evaluating higher-level audio-visual quality in scene-level generation."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
