{"record": {"aliases": [], "artifacts": [{"id": "arxiv:2606.08415", "kind": "paper", "url": "https://arxiv.org/abs/2606.08415"}, {"id": "gh:nju-link/covebench", "kind": "repo", "url": "https://github.com/NJU-LINK/CoVEBench"}], "authors": ["Jiangtao Wu", "Jiaming Wang", "Yiwen He", "Yuanxing Zhang", "Shihao Li", "Dunyuan Liu", "Xuedong Zhao", "Jialu Chen", "Zekun Moore Wang", "Jiaheng Liu"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2606.08415", "source_url": "https://arxiv.org/abs/2606.08415"}, "categories": ["General AI", "Vision & 3D", "cs.CV"], "description": {"en": "CoVEBench evaluates compositional video editing with 416 source videos, 626 multi-point instructions, and 9,990 checklist items, using MLLM judges and objective metrics."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2606.08415", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2606.08415", "title": "CoVEBench"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "institutions": ["Nanjing University", "Kuaishou (China)"], "institutions_reference": {"basis": "openalex_linked_institutions", "doi": "10.48550/arxiv.2606.08415", "retrieved_at": "2026-11-04", "source_key": "claire-radar:2606.08415", "source_url": "https://openalex.org/W7163986262"}, "key": "claire-radar:2606.08415", "modality": null, "name": "CoVEBench", "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_covebench_94152ad0", "origin_source": "arxiv", "recognition_confidence": "1.0", "record_sha256": "9c4397d22f8463c996c4446706de5e2f332a8c6f407e432eee49523d01f585a8", "relation": "introduces", "review_state": "unreviewed", "source_url": "https://arxiv.org/abs/2606.08415"}, "publisher": null, "released": "2026-06-07", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2606.08415", "source_url": "https://arxiv.org/abs/2606.08415"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2606-08415", "source": "claire_radar", "source_benchmark_id": "2606.08415", "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/NJU-LINK/CoVEBench", "githubScope": "benchmark_repo", "githubStars": 19, "hfDailySubmittedAt": null, "hfDataset": null, "hfDatasetDownloads": null, "hfDatasetLikes": null, "hfPaperUpvotes": 53, "hfPaperUrl": "https://huggingface.co/papers/2606.08415", "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": "2026-09-25", "state": "stale"}}, "status": "fresh"}, "capabilities": [], "confidence": "High", "construction": "Transform Existing", "constructionDetail": "Unknown — the indexer does not infer construction details without explicit source evidence.", "copyGeneration": {"generatedAt": "2026-08-20T15:22:10.044731Z", "inputHash": "24827a5a6257e03926cf514f6bd049e46bdc32156cb1b47d4448e6f561c11219", "model": "deepseek-v4-flash"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "CoVEBench evaluates compositional video editing with 416 source videos, 626 multi-point instructions, and 9,990 checklist items, using MLLM judges and objective metrics.", "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": "To address this gap, we introduce CoVEBench, a compositional video editing benchmark comprising 416 curated source videos, 626 multi-point editing instructions, and 9,990 fine-grained checklist items."}, "familyId": "bmf_9e4ba492c102", "firstSeenAt": "2026-08-19", "githubIndex": {"candidateCount": 0, "indexedAt": "2026-09-17T04:40:34.111665+00:00", "repositories": [{"evidence": [{"basis": "existing-code-link", "linkedUrl": "https://github.com/NJU-LINK/CoVEBench", "url": "https://github.com/NJU-LINK/CoVEBench"}, {"basis": "existing-metric-association", "linkedUrl": "https://github.com/NJU-LINK/CoVEBench", "url": "https://huggingface.co/papers/2606.08415"}, {"basis": "huggingface-paper-association", "linkedUrl": "https://github.com/NJU-LINK/CoVEBench", "url": "https://huggingface.co/api/papers/2606.08415"}], "scope": "benchmark_repo", "url": "https://github.com/NJU-LINK/CoVEBench", "verification": "source-linked"}], "status": "linked"}, "heat": null, "id": "bm_covebench_94152ad0", "indexedAt": "2026-08-19T11:05:13.395391Z", "industrySectors": [], "links": {"code": "https://github.com/NJU-LINK/CoVEBench", "data": null, "hfPaper": "https://huggingface.co/papers/2606.08415", "paper": "https://arxiv.org/abs/2606.08415", "pdf": "https://arxiv.org/pdf/2606.08415", "project": null, "report": "https://arxiv.org/abs/2606.08415"}, "metrics": [], "motivation": "While recent text-guided video editing models excel at elementary tasks (e.g., style transfer, object insertion), real-world user requests are highly compositional.", "name": "CoVEBench", "oneLine": "CoVEBench evaluates compositional video editing with 416 source videos, 626 multi-point instructions, and 9,990 checklist items, using MLLM judges and objective metrics.", "paperTitle": "CoVEBench: Can Video Editing Models Handle Complex Instructions?", "primaryDomain": "General AI", "publication": {"evidence": null, "evidenceLevel": "unverified", "evidenceUrl": "https://arxiv.org/abs/2606.08415", "source": "arxiv-metadata", "status": "unverified", "venue": null, "verifiedAt": "2026-10-14T01:46:58.058199Z"}, "publications": [], "publishers": [{"name": "NJU-LINK Team", "organizationType": "academic-lab", "role": "benchmark-publisher", "sourceUrl": "https://github.com/NJU-LINK/CoVEBench"}], "ranking": {}, "readiness": "Runnable", "recognitionConfidence": 1.0, "relation": "introduces", "releasedAt": "2026-06-07", "source": {"authors": ["Jiangtao Wu", "Jiaming Wang", "Yiwen He", "Yuanxing Zhang", "Shihao Li", "Dunyuan Liu", "Xuedong Zhao", "Jialu Chen", "Zekun Moore Wang", "Jiaheng Liu"], "categories": ["cs.CV", "cs.AI"], "id": "2606.08415", "title": "CoVEBench: Can Video Editing Models Handle Complex Instructions?", "type": "arxiv", "url": "https://arxiv.org/abs/2606.08415"}, "sourceUpdatedAt": "2026-06-10T04:17:35Z", "topics": ["cs.CV"], "venueAttempts": [], "whyItMatters": "Realistic video editing requires handling multiple coupled edits; CoVEBench provides a diagnostic testbed to reveal failures in complex instruction compliance and preservation."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
