{"record": {"aliases": [], "artifacts": [{"id": "arxiv:2606.01848", "kind": "paper", "url": "https://arxiv.org/abs/2606.01848"}, {"id": "gh:wukui-muc/rescuebench", "kind": "repo", "url": "https://github.com/wukui-muc/RescueBench"}], "authors": ["Kui Wu", "Beiyu Guo", "Hao Chen", "ShuHang Xu", "Yuling Li", "Yongdan Zeng", "Zhoujun Li", "Yizhou Wang", "Fangwei Zhong"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2606.01848", "source_url": "https://arxiv.org/abs/2606.01848"}, "categories": ["Multimodal", "Robotics & Embodied AI"], "description": {"en": "RescueBench evaluates embodied search-and-rescue agents in simulated photo-realistic environments. It comprises a four-stage pipeline: multimodal exploration, target rescue, memory-guided return, and final handoff. Five difficulty levels vary environmental complexity, clue ambiguity, and spatial hierarchy. Automatic episode generation and annotation support scalable evaluation. A unified benchmark framework and runner scripts provide standardized scoring."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2606.01848", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2606.01848", "title": "RescueBench"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "institutions": ["Beihang University", "City University of Macau", "Beijing Normal University", "Peking University"], "institutions_reference": {"basis": "openalex_linked_institutions", "doi": "10.48550/arxiv.2606.01848", "retrieved_at": "2026-11-04", "source_key": "claire-radar:2606.01848", "source_url": "https://openalex.org/W7163159293"}, "key": "claire-radar:2606.01848", "modality": null, "name": "RescueBench", "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_rescuebench_f2263257", "origin_source": "arxiv", "recognition_confidence": "1.0", "record_sha256": "cf962bbef34736b9c7d8167b8c1415f76a958cf40988231b5059f21aba59b5d0", "relation": "introduces", "review_state": "unreviewed", "source_url": "https://arxiv.org/abs/2606.01848"}, "publisher": null, "released": "2026-06-01", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2606.01848", "source_url": "https://arxiv.org/abs/2606.01848"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2606-01848", "source": "claire_radar", "source_benchmark_id": "2606.01848", "source_metadata": {"claire_radar": {"adoption30d": null, "aliases": [], "annotation": "Unknown", "applicationDomains": ["Robotics & Embodied AI"], "area": "Robotics & Embodied AI", "attention": {"asOf": "2026-10-05", "attemptedAt": "2026-10-04T15:31:48.030916Z", "githubRepo": "https://github.com/wukui-muc/RescueBench", "githubScope": "benchmark_repo", "githubStars": 8, "hfDailySubmittedAt": null, "hfDataset": null, "hfDatasetDownloads": null, "hfDatasetLikes": null, "hfPaperUpvotes": null, "hfPaperUrl": "https://huggingface.co/papers/2606.01848", "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", "state": "unavailable"}}, "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": "64a76691b682d95054288a0a3cf96cc73246b329fb3ee052d76fc4361bf07ea3", "model": "deepseek-v4-flash"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "RescueBench evaluates embodied search-and-rescue agents in simulated photo-realistic environments. It comprises a four-stage pipeline: multimodal exploration, target rescue, memory-guided return, and final handoff. Five difficulty levels vary environmental complexity, clue ambiguity, and spatial hierarchy. Automatic episode generation and annotation support scalable evaluation. A unified benchmark framework and runner scripts provide standardized scoring.", "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", "public artifact URL"], "snippet": "We introduce RescueBench, a photo-realistic diagnostic benchmark that instantiates SAR as a four-stage pipeline: multimodal exploration, target rescue, memory-guided return, and final handoff."}, "familyId": "bmf_6b6323c23009", "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/wukui-muc/RescueBench", "url": "https://github.com/wukui-muc/RescueBench"}, {"basis": "existing-metric-association", "linkedUrl": "https://github.com/wukui-muc/RescueBench", "url": "https://huggingface.co/papers/2606.01848"}, {"basis": "linked-from-report", "linkedUrl": "https://github.com/wukui-muc/RescueBench", "url": "https://arxiv.org/abs/2606.01848"}], "scope": "benchmark_repo", "url": "https://github.com/wukui-muc/RescueBench", "verification": "source-linked"}], "status": "linked"}, "heat": null, "id": "bm_rescuebench_f2263257", "indexedAt": "2026-08-19T11:05:13.395391Z", "industrySectors": ["Robotics"], "links": {"code": "https://github.com/wukui-muc/RescueBench", "data": null, "hfPaper": "https://huggingface.co/papers/2606.01848", "paper": "https://arxiv.org/abs/2606.01848", "pdf": "https://arxiv.org/pdf/2606.01848", "project": null, "report": "https://arxiv.org/abs/2606.01848"}, "metrics": [], "motivation": "Search-and-rescue (SAR) requires embodied agents to explore unfamiliar environments under multimodal uncertainty, perform multi-stage interactions, and retrieve spatial memory over long horizons.", "name": "RescueBench", "oneLine": "RescueBench evaluates embodied search-and-rescue agents in simulated photo-realistic environments. It comprises a four-stage pipeline: multimodal exploration, target rescue, memory-guided return, and final handoff. Five difficulty levels vary environmental complexity, clue ambiguity, and spatial hierarchy. Automatic episode generation and annotation support scalable evaluation. A unified benchmark framework and runner scripts provide standardized scoring.", "paperTitle": "RescueBench: Can Embodied Agents Save Lives in the Wild ?", "primaryDomain": "Robotics & Embodied AI", "publication": {"evidence": null, "evidenceLevel": "unverified", "evidenceUrl": "https://arxiv.org/abs/2606.01848", "source": "arxiv-metadata", "status": "unverified", "venue": null, "verifiedAt": "2026-10-14T01:46:58.058199Z"}, "publications": [], "ranking": {}, "readiness": "Runnable", "recognitionConfidence": 1.0, "relation": "introduces", "releasedAt": "2026-06-01", "source": {"authors": ["Kui Wu", "Beiyu Guo", "Hao Chen", "ShuHang Xu", "Yuling Li", "Yongdan Zeng", "Zhoujun Li", "Yizhou Wang", "Fangwei Zhong"], "categories": ["cs.CV"], "id": "2606.01848", "title": "RescueBench: Can Embodied Agents Save Lives in the Wild ?", "type": "arxiv", "url": "https://arxiv.org/abs/2606.01848"}, "sourceUpdatedAt": "2026-06-01T07:59:54Z", "topics": ["Multimodal"], "venueAttempts": [], "whyItMatters": "Search-and-rescue benchmarks typically test capabilities in isolation. RescueBench addresses the gap of composite workflows where failures may compound across stages. It provides stage-level diagnostics to identify bottlenecks (e.g., exploration, memory) separately from end-to-end performance, helping practitioners target improvements in embodied agents for realistic rescue tasks."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
