{"record": {"aliases": [], "artifacts": [{"id": "arxiv:2606.05665", "kind": "paper", "url": "https://arxiv.org/abs/2606.05665"}], "authors": ["Tao Liu", "Leela Krishna", "Gouti Pavan Kumar", "Sreeja K", "Vishav Garg"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2606.05665", "source_url": "https://arxiv.org/abs/2606.05665"}, "categories": ["General AI", "Safety", "Vision & 3D"], "description": {"en": "V2V-Bench evaluates video-to-video generation models across 11 dimensions in five categories: temporal alignment, structural fidelity, transformation quality, video quality, and semantic alignment. It pairs source videos with editing tasks and scores models on these dimensions."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2606.05665", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2606.05665", "title": "V2V-Bench"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "key": "claire-radar:2606.05665", "modality": null, "name": "V2V-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_v2v-bench_86c97c1d", "origin_source": "arxiv", "recognition_confidence": "1.0", "record_sha256": "9ffe6895115eb4338cea0e43e25f7255f8507d775a6928bfd865c3b5aa3bc520", "relation": "introduces", "review_state": "unreviewed", "source_url": "https://arxiv.org/abs/2606.05665"}, "publisher": null, "released": "2026-06-04", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2606.05665", "source_url": "https://arxiv.org/abs/2606.05665"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2606-05665", "source": "claire_radar", "source_benchmark_id": "2606.05665", "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": 0, "hfPaperUrl": "https://huggingface.co/papers/2606.05665", "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", "lastSuccessfulDate": "2026-09-25", "state": "stale"}}, "status": "stale"}, "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": "edc99abe4a50f224bdc676a072a8bb8d87646cfc5062ceff8091038fb55b60a5", "model": "deepseek-v4-flash"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "V2V-Bench evaluates video-to-video generation models across 11 dimensions in five categories: temporal alignment, structural fidelity, transformation quality, video quality, and semantic alignment. It pairs source videos with editing tasks and scores models on these dimensions.", "domainCuration": {"method": "rules-v1", "state": "auto"}, "evidence": {"reasonCodes": ["coined title prefix ending in Bench or Benchmark", "benchmark release stated in one sentence", "evaluation protocol evidence"], "snippet": "We introduce V2V-Bench, a 11-dimension benchmark organized into five categories: temporal alignment, structural fidelity, transformation quality, video quality, and semantic alignment."}, "familyId": "bmf_6f509f52ee80", "firstSeenAt": "2026-08-19", "githubIndex": {"candidateCount": 0, "indexedAt": "2026-09-17T04:40:34.111665+00:00", "repositories": [], "status": "not-found"}, "heat": null, "id": "bm_v2v-bench_86c97c1d", "indexedAt": "2026-08-19T11:05:13.395391Z", "industrySectors": [], "links": {"code": null, "data": null, "hfPaper": "https://huggingface.co/papers/2606.05665", "paper": "https://arxiv.org/abs/2606.05665", "pdf": "https://arxiv.org/pdf/2606.05665", "project": null, "report": "https://arxiv.org/abs/2606.05665"}, "metrics": [], "motivation": "Video-to-video (V2V) generation is difficult to evaluate because outputs must both follow editing instructions and preserve frame-level correspondence with the source video, which existing T2V and I2V metrics do not capture.", "name": "V2V-Bench", "oneLine": "V2V-Bench evaluates video-to-video generation models across 11 dimensions in five categories: temporal alignment, structural fidelity, transformation quality, video quality, and semantic alignment. It pairs source videos with editing tasks and scores models on these dimensions.", "paperTitle": "V2V-Bench: A Comprehensive Benchmark for Video-to-Video Generation Evaluation", "primaryDomain": "General AI", "publication": {"evidence": "Accepted at ICML 2026 workshop", "evidenceLevel": "author-claim", "evidenceUrl": "https://arxiv.org/abs/2606.05665", "source": "arxiv-comments", "status": "acceptance_claimed", "venue": "ICML 2026 workshop", "verifiedAt": "2026-10-14T01:46:58.058199Z"}, "publications": [], "ranking": {}, "readiness": "Paper only", "recognitionConfidence": 1.0, "relation": "introduces", "releasedAt": "2026-06-04", "source": {"authors": ["Tao Liu", "Leela Krishna", "Gouti Pavan Kumar", "Sreeja K", "Vishav Garg"], "categories": ["cs.CV"], "id": "2606.05665", "title": "V2V-Bench: A Comprehensive Benchmark for Video-to-Video Generation Evaluation", "type": "arxiv", "url": "https://arxiv.org/abs/2606.05665"}, "sourceUpdatedAt": "2026-06-04T03:48:42Z", "topics": ["Safety"], "venueAttempts": [{"decisionRaw": "Accepted at ICML 2026 workshop", "evidence": [{"level": "author-claim", "observedAt": "2026-10-14T01:46:58.058199Z", "rawValue": "Accepted at ICML 2026 workshop", "sourceType": "arxiv-comments", "sourceUrl": "https://arxiv.org/abs/2606.05665"}], "reviewStatus": "accepted", "venueName": "ICML 2026 workshop"}], "whyItMatters": "Existing T2V and I2V metrics do not capture the joint requirements of instruction following and frame-level correspondence in V2V generation. A dedicated benchmark with human-correlated scoring could support model selection for V2V applications."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
