{"record": {"aliases": [], "artifacts": [{"id": "arxiv:2606.21949", "kind": "paper", "url": "https://arxiv.org/abs/2606.21949"}, {"id": "gh:xlchen0205/capricorn-1k", "kind": "repo", "url": "https://github.com/xlchen0205/CapRiCorn-1K"}], "authors": ["Xinlong Chen", "Jiafu Tang", "Yue Ding", "Yizhuo Jia", "Bozhou Li", "Bohan Zeng", "Yang Shi", "Shihao Li", "Yiyan Ji", "Qiang Liu", "Weihong Lin", "Yuanxing Zhang", "Pengfei Wan", "Liang Wang", "Tieniu Tan"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2606.21949", "source_url": "https://arxiv.org/abs/2606.21949"}, "categories": ["General AI", "Vision & 3D", "cs.CV"], "description": {"en": "CapRiCorn-1K evaluates video captioning quality and subject referential consistency across long videos (15s-10min) with audiovisual and visual-only settings. It uses LLM judge to compute accuracy, coverage, and referential consistency metrics based on manual annotations."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2606.21949", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2606.21949", "title": "CapRiCorn-1K"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "key": "claire-radar:2606.21949", "modality": null, "name": "CapRiCorn-1K", "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_capricorn-1k_86185f2f", "origin_source": "arxiv", "recognition_confidence": "1.0", "record_sha256": "a65615789b75a821dd8a8fa709dd74cae1c930b327d3a12b3e30eb3fe65b63d9", "relation": "introduces", "review_state": "unreviewed", "source_url": "https://arxiv.org/abs/2606.21949"}, "publisher": null, "released": "2026-06-20", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2606.21949", "source_url": "https://arxiv.org/abs/2606.21949"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2606-21949", "source": "claire_radar", "source_benchmark_id": "2606.21949", "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/xlchen0205/CapRiCorn-1K", "githubScope": "benchmark_repo", "githubStars": 0, "hfDailySubmittedAt": null, "hfDataset": null, "hfDatasetDownloads": null, "hfDatasetLikes": null, "hfPaperUpvotes": 0, "hfPaperUrl": "https://huggingface.co/papers/2606.21949", "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": "2026-10-04", "state": "stale"}}, "status": "fresh"}, "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": "41a9d510b9fcfdd080442ff8f92514749267c03bfdd2dedd97a3e59949083339", "model": "deepseek-v4-flash"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "CapRiCorn-1K evaluates video captioning quality and subject referential consistency across long videos (15s-10min) with audiovisual and visual-only settings. It uses LLM judge to compute accuracy, coverage, and referential consistency metrics based on manual annotations.", "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", "public artifact URL"], "snippet": "To bridge this gap, we propose CapRiCorn-1K, a comprehensive benchmark designed to evaluate both video captioning quality and subject referential consistency across long temporal horizons and diverse video domains."}, "familyId": "bmf_ea863771b451", "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/xlchen0205/CapRiCorn-1K", "url": "https://github.com/xlchen0205/CapRiCorn-1K"}, {"basis": "existing-metric-association", "linkedUrl": "https://github.com/xlchen0205/CapRiCorn-1K", "url": "https://huggingface.co/papers/2606.21949"}, {"basis": "linked-from-report", "linkedUrl": "https://github.com/xlchen0205/CapRiCorn-1K", "url": "https://arxiv.org/abs/2606.21949"}], "scope": "benchmark_repo", "url": "https://github.com/xlchen0205/CapRiCorn-1K", "verification": "source-linked"}], "status": "linked"}, "heat": null, "id": "bm_capricorn-1k_86185f2f", "indexedAt": "2026-08-19T11:05:13.395391Z", "industrySectors": [], "links": {"code": "https://github.com/xlchen0205/CapRiCorn-1K", "data": null, "hfPaper": "https://huggingface.co/papers/2606.21949", "paper": "https://arxiv.org/abs/2606.21949", "pdf": "https://arxiv.org/pdf/2606.21949", "project": null, "report": "https://arxiv.org/abs/2606.21949"}, "metrics": [], "motivation": "Accurate and comprehensive video captions with consistent subject references are critical for downstream understanding and generation tasks.", "name": "CapRiCorn-1K", "oneLine": "CapRiCorn-1K evaluates video captioning quality and subject referential consistency across long videos (15s-10min) with audiovisual and visual-only settings. It uses LLM judge to compute accuracy, coverage, and referential consistency metrics based on manual annotations.", "paperTitle": "CapRiCorn-1K: A Comprehensive Benchmark for Video Captioning and Subject Referential Consistency Across Temporal Scales", "primaryDomain": "General AI", "publication": {"evidence": null, "evidenceLevel": "unverified", "evidenceUrl": "https://arxiv.org/abs/2606.21949", "source": "arxiv-metadata", "status": "unverified", "venue": null, "verifiedAt": "2026-10-14T01:46:58.058199Z"}, "publications": [], "publishers": [{"name": "CapRiCorn-1K Team", "organizationType": "academic-lab", "role": "benchmark-publisher", "sourceUrl": "https://github.com/xlchen0205/CapRiCorn-1K"}], "ranking": {}, "readiness": "Runnable", "recognitionConfidence": 1.0, "relation": "introduces", "releasedAt": "2026-06-20", "source": {"authors": ["Xinlong Chen", "Jiafu Tang", "Yue Ding", "Yizhuo Jia", "Bozhou Li", "Bohan Zeng", "Yang Shi", "Shihao Li", "Yiyan Ji", "Qiang Liu", "Weihong Lin", "Yuanxing Zhang", "Pengfei Wan", "Liang Wang", "Tieniu Tan"], "categories": ["cs.CV", "cs.CL"], "id": "2606.21949", "title": "CapRiCorn-1K: A Comprehensive Benchmark for Video Captioning and Subject Referential Consistency Across Temporal Scales", "type": "arxiv", "url": "https://arxiv.org/abs/2606.21949"}, "sourceUpdatedAt": "2026-06-20T08:37:32Z", "topics": ["cs.CV"], "venueAttempts": [], "whyItMatters": "Existing benchmarks focus on short videos and overall caption quality, lacking evaluation of subject referential consistency over long horizons. CapRiCorn-1K's metrics correlate with downstream understanding and generation performance, offering practical value for selecting captioning models."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
