{"record": {"aliases": [], "artifacts": [{"id": "arxiv:2606.08572", "kind": "paper", "url": "https://arxiv.org/abs/2606.08572"}, {"id": "gh:nju-link/omnicap-if", "kind": "repo", "url": "https://github.com/NJU-LINK/omnicap-if"}], "authors": ["Jiahao Wang", "An Ping", "Yanghai Wang", "Yuanxing Zhang", "Shihao Li", "Hanyan Bian", "Yichi Ren", "Yize Zhang", "Han Wang", "Haowen Chen", "Junze Li", "Jiaqi Wang", "Yiyang Hu", "Zhuze Xu", "Zijie Zhang", "Jiaheng Liu"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2606.08572", "source_url": "https://arxiv.org/abs/2606.08572"}, "categories": ["General AI", "Vision & 3D", "cs.CV"], "description": {"en": "OmniCap-IF evaluates instruction following in omni-modal video captioning with 50 constraint types, 1,920 samples, and checklist-based scoring for format and content correctness."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2606.08572", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2606.08572", "title": "OmniCap-IF"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "key": "claire-radar:2606.08572", "modality": null, "name": "OmniCap-IF", "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_omnicap-if_98c29629", "origin_source": "arxiv", "recognition_confidence": "1.0", "record_sha256": "6b3fa68354fec1caca6c3758ab7c569a0f8fed06084311c0819e09101ec03fbd", "relation": "introduces", "review_state": "unreviewed", "source_url": "https://arxiv.org/abs/2606.08572"}, "publisher": null, "released": "2026-06-07", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2606.08572", "source_url": "https://arxiv.org/abs/2606.08572"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2606-08572", "source": "claire_radar", "source_benchmark_id": "2606.08572", "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/omnicap-if", "githubScope": "benchmark_repo", "githubStars": 11, "hfDailySubmittedAt": null, "hfDataset": null, "hfDatasetDownloads": null, "hfDatasetLikes": null, "hfPaperUpvotes": 14, "hfPaperUrl": "https://huggingface.co/papers/2606.08572", "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": "8c8ffb534a25992679278a1e0d6143005bf114ca3eed4f0946c1a9b4ef0d8164", "model": "deepseek-v4-flash"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "OmniCap-IF evaluates instruction following in omni-modal video captioning with 50 constraint types, 1,920 samples, and checklist-based scoring for format and content correctness.", "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 bridge this gap, we introduce OmniCap-IF, the first comprehensive benchmark specifically designed to evaluate instruction-following capabilities in omni-modal captioning."}, "familyId": "bmf_3e95cd9aa687", "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/omnicap-if", "url": "https://github.com/NJU-LINK/omnicap-if"}, {"basis": "existing-metric-association", "linkedUrl": "https://github.com/NJU-LINK/omnicap-if", "url": "https://huggingface.co/papers/2606.08572"}], "scope": "benchmark_repo", "url": "https://github.com/NJU-LINK/omnicap-if", "verification": "source-linked"}], "status": "linked"}, "heat": null, "id": "bm_omnicap-if_98c29629", "indexedAt": "2026-08-19T11:05:13.395391Z", "industrySectors": [], "links": {"code": "https://github.com/NJU-LINK/omnicap-if", "data": null, "hfPaper": "https://huggingface.co/papers/2606.08572", "paper": "https://arxiv.org/abs/2606.08572", "pdf": "https://arxiv.org/pdf/2606.08572", "project": null, "report": "https://arxiv.org/abs/2606.08572"}, "metrics": [], "motivation": "While Omni-modal Large Language Models (OLLMs) have demonstrated impressive capabilities in jointly processing audio and visual streams, their ability to strictly adhere to complex, multi-faceted user instructions remains largely unexplored.", "name": "OmniCap-IF", "oneLine": "OmniCap-IF evaluates instruction following in omni-modal video captioning with 50 constraint types, 1,920 samples, and checklist-based scoring for format and content correctness.", "paperTitle": "OmniCap-IF: Benchmarking and Improving Instruction Following Abilities for Omni-Video Captioning", "primaryDomain": "General AI", "publication": {"evidence": null, "evidenceLevel": "unverified", "evidenceUrl": "https://arxiv.org/abs/2606.08572", "source": "arxiv-metadata", "status": "unverified", "venue": null, "verifiedAt": "2026-10-14T01:46:58.058199Z"}, "publications": [], "publishers": [{"name": "NJU-LINK", "organizationType": "academic-lab", "role": "benchmark-publisher", "sourceUrl": "https://github.com/NJU-LINK/omnicap-if"}], "ranking": {}, "readiness": "Runnable", "recognitionConfidence": 1.0, "relation": "introduces", "releasedAt": "2026-06-07", "source": {"authors": ["Jiahao Wang", "An Ping", "Yanghai Wang", "Yuanxing Zhang", "Shihao Li", "Hanyan Bian", "Yichi Ren", "Yize Zhang", "Han Wang", "Haowen Chen", "Junze Li", "Jiaqi Wang", "Yiyang Hu", "Zhuze Xu", "Zijie Zhang", "Jiaheng Liu"], "categories": ["cs.CV"], "id": "2606.08572", "title": "OmniCap-IF: Benchmarking and Improving Instruction Following Abilities for Omni-Video Captioning", "type": "arxiv", "url": "https://arxiv.org/abs/2606.08572"}, "sourceUpdatedAt": "2026-06-07T11:03:33Z", "topics": ["cs.CV"], "venueAttempts": [], "whyItMatters": "Existing benchmarks miss the interplay of audio-visual and user constraints; OmniCap-IF provides a fine-grained evaluation to expose format-content tradeoffs and drive improvements."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
