{"record": {"aliases": [], "artifacts": [{"id": "arxiv:2608.18607", "kind": "paper", "url": "https://arxiv.org/abs/2608.18607"}, {"id": "gh:sharelab-sii/va-judger", "kind": "repo", "url": "https://github.com/ShareLab-SII/VA-Judger"}], "categories": ["General AI", "Vision & 3D", "cs.CV"], "description": {"en": "VA-Judger-Bench evaluates reward models for joint video-audio generation. It contains paired comparisons of generated video-audio samples with human preference labels, covering in-domain and out-of-domain model outputs."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2608.18607", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2608.18607", "title": "VA-Judger-Bench"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "key": "claire-radar:2608.18607", "modality": null, "name": "VA-Judger-Bench", "openness": {"code_license": null, "data_license": null, "evidence": [], "status": "unknown"}, "provenance": {"confidence": "Medium", "crawl_bundle": "claire_radar_2026-10-14", "crawled_at": "2026-10-14T09:44:51Z", "data_status": "primary-source-indexed", "display_eligible": "true", "origin_record_id": "bm_va-judger-bench_7e8bd378", "origin_source": "arxiv", "recognition_confidence": "0.55", "record_sha256": "907ecadfd8934ad53fad392772d14bf2b9128966707644f9fc978c7f0abd9295", "relation": "evaluates_only", "review_model": "deepseek-v4-flash", "review_state": "ai-reviewed", "reviewed_at": "2026-08-20T09:45:58.504563Z", "source_url": "https://arxiv.org/abs/2608.18607"}, "publisher": null, "released": "2026-08-19", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2608.18607", "source_url": "https://arxiv.org/abs/2608.18607"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2608-18607", "source": "claire_radar", "source_benchmark_id": "2608.18607", "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": "https://github.com/ShareLab-SII/VA-Judger", "githubScope": "benchmark_repo", "githubStars": 64, "hfDailySubmittedAt": null, "hfDataset": null, "hfDatasetDownloads": null, "hfDatasetLikes": null, "hfPaperUpvotes": 12, "hfPaperUrl": "https://huggingface.co/papers/2608.18607", "observedAt": "2026-10-04T15:31:48.030916Z", "signalStatus": {"githubStars": {"observedAt": "2026-10-04T15:31:48.030916Z", "state": "fresh"}, "hfDatasetDownloads": {"state": "not_applicable"}, "hfDatasetLikes": {"state": "not_applicable"}, "hfPaperUpvotes": {"checkedAt": "2026-10-04T15:31:48.030916Z", "lastSuccessfulDate": "2026-09-30", "state": "stale"}}, "status": "fresh"}, "capabilities": [], "confidence": "Medium", "construction": "Unknown", "constructionDetail": "Unknown — the indexer does not infer construction details without explicit source evidence.", "copyGeneration": {"generatedAt": "2026-08-20T15:22:10.044731Z", "inputHash": "f04b475f5407961dea11eea659358dca79197764b37006841a6c009cabb9b383", "model": "deepseek-v4-flash"}, "curation": {"decisionReason": "VA-Judger-Bench is a named benchmark with a defined evaluation protocol, publicly released data/checkpoints, and clear scoring contract for comparing reward models.", "model": "deepseek-v4-flash", "reviewedAt": "2026-08-20T09:45:58.504563Z", "state": "ai-reviewed"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "VA-Judger-Bench evaluates reward models for joint video-audio generation. It contains paired comparisons of generated video-audio samples with human preference labels, covering in-domain and out-of-domain model outputs.", "displayEligible": true, "domainCuration": {"method": "rules-v1", "state": "auto"}, "evaluationMode": "public_reusable", "evidence": {"reasonCodes": ["exact named benchmark artifact released in abstract", "evaluation protocol evidence", "public artifact URL", "named entity is a model, method, framework, or dataset used in evaluation"], "snippet": "We also introduce the VA-Judger-Bench benchmark with both in-domain and out-of-domain model comparisons to evaluate whether reward models truly align with human preferences."}, "familyId": "bmf_52e48663243f", "firstSeenAt": "2026-08-20", "heat": null, "id": "bm_va-judger-bench_7e8bd378", "indexedAt": "2026-08-20T09:44:47.308583Z", "industrySectors": [], "links": {"code": "https://github.com/ShareLab-SII/VA-Judger", "data": null, "paper": "https://arxiv.org/abs/2608.18607", "pdf": "https://arxiv.org/pdf/2608.18607", "project": null, "report": "https://arxiv.org/abs/2608.18607"}, "metrics": [], "motivation": "Using reinforcement learning to post-train joint video-audio generation models requires a reward signal.", "name": "VA-Judger-Bench", "oneLine": "VA-Judger-Bench evaluates reward models for joint video-audio generation. It contains paired comparisons of generated video-audio samples with human preference labels, covering in-domain and out-of-domain model outputs.", "paperTitle": "VA-Judger: Reward Modeling from Human Preference Feedback for Joint Video-Audio Generation", "primaryDomain": "General AI", "publication": {"evidence": null, "evidenceLevel": "unverified", "evidenceUrl": "https://arxiv.org/abs/2608.18607", "source": "arxiv-metadata", "status": "unverified", "venue": null, "verifiedAt": "2026-08-20T09:44:47.308583Z"}, "publications": [], "publishers": [{"name": "ShareLab-SII", "organizationType": "academic-lab", "role": "benchmark-publisher", "sourceUrl": "https://github.com/ShareLab-SII/VA-Judger"}], "ranking": {"90d": {"confidence": "Medium", "coverage": 0.7, "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": 0.5572056612399781, "value": 64}, "hfDatasetDownloads": {"normalization": "log-max", "normalized": null, "value": null}, "hfPaperUpvotes": {"normalization": "log-max", "normalized": 0.4532127250650379, "value": 12}}, "confidence": "Medium", "coverage": 0.7, "rank": 47, "score": 52}, "method": "attention-ranking-v13", "rank": 47, "score": 52}}, "readiness": "Runnable", "recognitionConfidence": 0.55, "relation": "evaluates_only", "releasedAt": "2026-08-19", "source": {"authors": ["Yinming Huang", "Shuyuan Tu", "Xi Yan", "Zihan Yang", "Jianhua Han", "Xu Hang", "Yu-Gang Jiang", "Zuxuan Wu"], "categories": ["cs.CV"], "id": "2608.18607", "title": "VA-Judger: Reward Modeling from Human Preference Feedback for Joint Video-Audio Generation", "type": "arxiv", "url": "https://arxiv.org/abs/2608.18607"}, "sourceUpdatedAt": "2026-08-19T06:53:14Z", "topics": ["cs.CV"], "venueAttempts": [], "whyItMatters": "Existing metrics evaluate quality dimensions separately, missing semantic and temporal coherence. VA-Judger-Bench provides a benchmark to assess whether reward models align with human preferences for joint video-audio generation."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
