{"record": {"aliases": [], "artifacts": [{"id": "arxiv:2608.00012", "kind": "paper", "url": "https://arxiv.org/abs/2608.00012"}], "authors": ["Fengxiang Wang", "Qiuyang Yu", "Yueying Li", "Mingshuo Chen", "Chengchi Fei", "Kaiyi Xu", "Lixin Gu", "Wangxu Wei", "Junchao Gong", "Lipeng Ma", "Jiong Wang", "Fenghua Ling", "Wenlong Zhang", "Xue Yang", "Wenjing Yang", "Ben Fei", "Long Lan"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2608.00012", "source_url": "https://arxiv.org/abs/2608.00012"}, "categories": ["General AI", "Multimodal"], "description": {"en": "Obshazard-bench is a real-time benchmark for disaster intelligence, integrating raw satellite and ground-station data. It covers 8 disaster categories and 28 sub-categories across 60+ countries, with VQA samples and a three-stage evaluation taxonomy: Predictive Crisis Anticipation, Active Evolution Reasoning, and Multi-faceted Impact Quantification."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2608.00012", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2608.00012", "title": "Obshazard-bench"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "key": "claire-radar:2608.00012", "modality": null, "name": "Obshazard-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_obshazard-bench_aa08033a", "origin_source": "arxiv", "recognition_confidence": "1.0", "record_sha256": "50401839248505667eaf0ed5f9d3d635eda99910c04164eaf6f5f072bcc7e574", "relation": "introduces", "review_state": "unreviewed", "source_url": "https://arxiv.org/abs/2608.00012"}, "publisher": null, "released": "2026-06-24", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2608.00012", "source_url": "https://arxiv.org/abs/2608.00012"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2608-00012", "source": "claire_radar", "source_benchmark_id": "2608.00012", "source_metadata": {"claire_radar": {"adoption30d": null, "aliases": [], "annotation": "Unknown", "applicationDomains": ["General AI"], "area": "Multimodal", "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": null, "hfPaperUrl": "https://huggingface.co/papers/2608.00012", "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", "state": "unavailable"}}, "status": "unavailable"}, "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": "52071dd3e129915c8dee4b0baa8e5a5714fed0165eef5d197a01434a8343478e", "model": "deepseek-v4-flash"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "Obshazard-bench is a real-time benchmark for disaster intelligence, integrating raw satellite and ground-station data. It covers 8 disaster categories and 28 sub-categories across 60+ countries, with VQA samples and a three-stage evaluation taxonomy: Predictive Crisis Anticipation, Active Evolution Reasoning, and Multi-faceted Impact Quantification.", "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 bridge this gap, we introduce Obshazard-bench, a real-time, observation-driven benchmark for evaluating disaster intelligence in MLLMs."}, "familyId": "bmf_697b42360627", "firstSeenAt": "2026-08-19", "githubIndex": {"candidateCount": 0, "indexedAt": "2026-09-17T04:40:34.111665+00:00", "repositories": [], "status": "not-found"}, "heat": null, "id": "bm_obshazard-bench_aa08033a", "indexedAt": "2026-08-19T11:05:13.395391Z", "industrySectors": [], "links": {"code": null, "data": null, "hfPaper": "https://huggingface.co/papers/2608.00012", "paper": "https://arxiv.org/abs/2608.00012", "pdf": "https://arxiv.org/pdf/2608.00012", "project": null, "report": "https://arxiv.org/abs/2608.00012"}, "metrics": [], "motivation": "Multimodal Large Language Models (MLLMs) are increasingly used to interpret Earth observation data, yet their capability to support real-world disaster emergency response remains insufficiently evaluated.", "name": "Obshazard-bench", "oneLine": "Obshazard-bench is a real-time benchmark for disaster intelligence, integrating raw satellite and ground-station data. It covers 8 disaster categories and 28 sub-categories across 60+ countries, with VQA samples and a three-stage evaluation taxonomy: Predictive Crisis Anticipation, Active Evolution Reasoning, and Multi-faceted Impact Quantification.", "paperTitle": "Obshazard-bench: Benchmarking Multimodal Foundation Models for Real-Time Disaster Intelligence from Raw Earth Observation Streams", "primaryDomain": "General AI", "publication": {"evidence": null, "evidenceLevel": "unverified", "evidenceUrl": "https://arxiv.org/abs/2608.00012", "source": "arxiv-metadata", "status": "unverified", "venue": null, "verifiedAt": "2026-10-14T01:46:58.058199Z"}, "publications": [], "publishers": [{"name": "arXiv", "organizationType": "community", "role": "benchmark-publisher", "sourceUrl": "https://arxiv.org/abs/2608.00012"}], "ranking": {}, "readiness": "Paper only", "recognitionConfidence": 1.0, "relation": "introduces", "releasedAt": "2026-06-24", "source": {"authors": ["Fengxiang Wang", "Qiuyang Yu", "Yueying Li", "Mingshuo Chen", "Chengchi Fei", "Kaiyi Xu", "Lixin Gu", "Wangxu Wei", "Junchao Gong", "Lipeng Ma", "Jiong Wang", "Fenghua Ling", "Wenlong Zhang", "Xue Yang", "Wenjing Yang", "Ben Fei", "Long Lan"], "categories": ["cs.CL", "cs.AI", "cs.CY", "cs.LG"], "id": "2608.00012", "title": "Obshazard-bench: Benchmarking Multimodal Foundation Models for Real-Time Disaster Intelligence from Raw Earth Observation Streams", "type": "arxiv", "url": "https://arxiv.org/abs/2608.00012"}, "sourceUpdatedAt": "2026-06-24T07:51:49Z", "topics": ["Multimodal"], "venueAttempts": [], "whyItMatters": "Fills the gap in evaluating MLLMs for operational disaster response, which require real-time reasoning from raw observation streams. It provides a realistic testbed for assessing decision-support capabilities in evolving emergencies."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
