{"record": {"aliases": [], "artifacts": [{"id": "arxiv:2610.12417", "kind": "paper", "url": "https://arxiv.org/abs/2610.12417"}], "authors": ["Zheyu Fan", "Yue Zhang", "Mingkai Deng", "Kangrui Wang", "Qineng Wang", "Canyu Chen", "Jie Hao", "Xing Fan", "Chenlei Guo", "Eric P. Xing", "Mohit Bansal", "Manling Li"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2610.12417", "source_url": "https://arxiv.org/abs/2610.12417"}, "categories": ["Multimodal", "Reasoning", "Robotics & Embodied AI"], "description": {"en": "Multimodal large language models (MLLMs) struggle with spatial, embodied, physical, and temporal reasoning."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2610.12417", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2610.12417", "title": "WOVEN"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "key": "claire-radar:2610.12417", "modality": null, "name": "WOVEN", "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_woven_dfbe3ac8", "origin_source": "arxiv", "recognition_confidence": "0.85", "record_sha256": "83e5cfc6ebe3060591c02035620b0c38425debb5ea5e14ff6276f7bb02e9e4fc", "relation": "introduces", "review_state": "rule-auto-admitted", "reviewed_at": "2026-10-11T07:41:20.682221Z", "source_url": "https://arxiv.org/abs/2610.12417"}, "publisher": null, "released": "2026-10-08", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2610.12417", "source_url": "https://arxiv.org/abs/2610.12417"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2610-12417", "source": "claire_radar", "source_benchmark_id": "2610.12417", "source_metadata": {"claire_radar": {"adoption30d": null, "aliases": [], "annotation": "Unknown", "applicationDomains": ["Robotics & Embodied AI"], "area": "Multimodal", "capabilities": ["Reasoning"], "confidence": "Medium", "construction": "Unknown", "constructionDetail": "Unknown — the indexer does not infer construction details without explicit source evidence.", "curation": {"decisionReason": "Automatically admitted from an explicit named benchmark release with evaluation-protocol evidence at or above the configured publication-confidence threshold.", "policy": "source-evidence-v2", "reviewedAt": "2026-10-11T07:41:20.682221Z", "state": "rule-auto-admitted"}, "dataStatus": "primary-source-indexed", "demo": false, "displayEligible": true, "domainCuration": {"method": "rules-v1", "state": "auto"}, "evaluationMode": "public_reusable", "evidence": {"reasonCodes": ["exact coined title identity tied to benchmark evidence", "evaluation protocol evidence"], "snippet": "We therefore introduce WOVEN, a training source and benchmark for visual transition reasoning that organizes transition supervision by scene, action, and reasoning type, using diverse, realistic rollouts from video-pretrained generative models: 36,076 examples across 20 scene types, 5 action types, and 8 reasoning types."}, "familyId": "bmf_97d8b48aaf83", "firstSeenAt": "2026-10-09", "heat": null, "id": "bm_woven_dfbe3ac8", "indexedAt": "2026-10-11T07:41:00.871216Z", "industrySectors": ["Robotics"], "links": {"code": null, "data": null, "paper": "https://arxiv.org/abs/2610.12417", "pdf": "https://arxiv.org/pdf/2610.12417", "project": null, "report": "https://arxiv.org/abs/2610.12417"}, "metrics": [], "motivation": "Multimodal large language models (MLLMs) struggle with spatial, embodied, physical, and temporal reasoning.", "name": "WOVEN", "oneLine": "Multimodal large language models (MLLMs) struggle with spatial, embodied, physical, and temporal reasoning.", "paperTitle": "WOVEN: Weaving Visual World Modeling into Multimodal LLMs", "primaryDomain": "Robotics & Embodied AI", "publication": {"evidence": null, "evidenceLevel": "unverified", "evidenceUrl": "https://arxiv.org/abs/2610.12417", "source": "arxiv-metadata", "status": "unverified", "venue": null, "verifiedAt": "2026-10-14T01:46:58.058199Z"}, "publications": [], "readiness": "Paper only", "recognitionConfidence": 0.85, "relation": "introduces", "releasedAt": "2026-10-08", "source": {"authors": ["Zheyu Fan", "Yue Zhang", "Mingkai Deng", "Kangrui Wang", "Qineng Wang", "Canyu Chen", "Jie Hao", "Xing Fan", "Chenlei Guo", "Eric P. Xing", "Mohit Bansal", "Manling Li"], "categories": ["cs.CV", "cs.CL", "cs.LG"], "id": "2610.12417", "title": "WOVEN: Weaving Visual World Modeling into Multimodal LLMs", "type": "arxiv", "url": "https://arxiv.org/abs/2610.12417"}, "sourceUpdatedAt": "2026-10-08T17:52:03Z", "topics": ["Multimodal", "Reasoning"], "venueAttempts": []}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
