{"record": {"aliases": [], "artifacts": [{"id": "arxiv:2607.13653", "kind": "paper", "url": "https://arxiv.org/abs/2607.13653"}, {"id": "gh:internrobotics/real", "kind": "repo", "url": "https://github.com/InternRobotics/REAL"}], "authors": ["Boyu Mi", "Mengchen Ma", "Yifei Yao", "Xing Gao", "Junting Chen", "Yangzi Li", "Zihou Zhu", "Guohao Li", "Zhenfei Yin", "Tai Wang", "Yao Mu", "Jiangmiao Pang", "Hanqing Wang"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2607.13653", "source_url": "https://arxiv.org/abs/2607.13653"}, "categories": ["Robot manipulation", "Robotics", "Robotics & Embodied AI"], "description": {"en": "REAL-Bench evaluates vision-driven embodied agents in open-world mobile manipulation across 241 tasks spanning active exploration, visual distraction, articulated manipulation, and interactive disambiguation. The benchmark provides standardized task definitions and a simulator-based evaluation protocol."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2607.13653", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2607.13653", "title": "REAL-Bench"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "key": "claire-radar:2607.13653", "modality": null, "name": "REAL-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_real-bench_906823b4", "origin_source": "arxiv", "recognition_confidence": "1.0", "record_sha256": "9d1713adbfef5ce07799cb445bc892840ce95a62752a7d30070c33b02823ea6f", "relation": "introduces", "review_state": "unreviewed", "source_url": "https://arxiv.org/abs/2607.13653"}, "publisher": null, "released": "2026-07-15", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2607.13653", "source_url": "https://arxiv.org/abs/2607.13653"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2607-13653", "source": "claire_radar", "source_benchmark_id": "2607.13653", "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/InternRobotics/REAL", "githubScope": "benchmark_repo", "githubStars": 56, "hfDailySubmittedAt": null, "hfDataset": null, "hfDatasetDownloads": null, "hfDatasetLikes": null, "hfPaperUpvotes": 0, "hfPaperUrl": "https://huggingface.co/papers/2607.13653", "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": {"observedAt": "2026-10-04T15:31:48.030916Z", "state": "fresh"}}, "status": "fresh"}, "capabilities": ["Robot manipulation"], "confidence": "High", "construction": "Interactive Environment", "constructionDetail": "Unknown — the indexer does not infer construction details without explicit source evidence.", "copyGeneration": {"generatedAt": "2026-08-20T15:22:10.044731Z", "inputHash": "d87a8843dd9c4628b22e06c66c59b9940846649238afd448749ac5ba5b990da7", "model": "deepseek-v4-flash"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "REAL-Bench evaluates vision-driven embodied agents in open-world mobile manipulation across 241 tasks spanning active exploration, visual distraction, articulated manipulation, and interactive disambiguation. The benchmark provides standardized task definitions and a simulator-based evaluation protocol.", "domainCuration": {"method": "rules-v1", "state": "auto"}, "evidence": {"reasonCodes": ["exact named benchmark artifact released in abstract", "benchmark release stated in one sentence", "evaluation protocol evidence", "public artifact URL"], "snippet": "To comprehensively evaluate this approach, we introduce REAL-Bench, a benchmark spanning 241 tasks across active exploration, visual distraction, articulated manipulation, and interactive disambiguation."}, "familyId": "bmf_f0848f948867", "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/InternRobotics/REAL", "url": "https://github.com/InternRobotics/REAL"}, {"basis": "existing-metric-association", "linkedUrl": "https://github.com/InternRobotics/REAL", "url": "https://huggingface.co/papers/2607.13653"}, {"basis": "linked-from-report", "linkedUrl": "https://github.com/InternRobotics/REAL.", "url": "https://arxiv.org/abs/2607.13653"}, {"basis": "linked-from-report", "linkedUrl": "https://github.com/InternRobotics/REAL", "url": "https://arxiv.org/abs/2607.13653"}], "scope": "project_repo", "url": "https://github.com/InternRobotics/REAL", "verification": "source-linked"}], "status": "linked"}, "heat": null, "id": "bm_real-bench_906823b4", "indexedAt": "2026-08-19T11:05:13.395391Z", "industrySectors": ["Robotics"], "links": {"code": "https://github.com/InternRobotics/REAL", "data": null, "hfPaper": "https://huggingface.co/papers/2607.13653", "paper": "https://arxiv.org/abs/2607.13653", "pdf": "https://arxiv.org/pdf/2607.13653", "project": null, "report": "https://arxiv.org/abs/2607.13653"}, "metrics": [], "motivation": "Real-world deployment of embodied agents requires active exploration, visual grounding, and interactive intent disambiguation.", "name": "REAL-Bench", "oneLine": "REAL-Bench evaluates vision-driven embodied agents in open-world mobile manipulation across 241 tasks spanning active exploration, visual distraction, articulated manipulation, and interactive disambiguation. The benchmark provides standardized task definitions and a simulator-based evaluation protocol.", "paperTitle": "Exploratory, Communicative, and Deployable: Vision-Driven Embodied Agents for Open-World Mobile Manipulation", "primaryDomain": "Robotics & Embodied AI", "publication": {"evidence": "Accepted to ECCV 2026. 57 pages. Code available at https://github.com/InternRobotics/REAL", "evidenceLevel": "author-claim", "evidenceUrl": "https://arxiv.org/abs/2607.13653", "source": "arxiv-comments", "status": "acceptance_claimed", "venue": "ECCV 2026", "verifiedAt": "2026-10-14T01:46:58.058199Z"}, "publications": [], "publishers": [{"name": "InternRobotics", "organizationType": "academic-lab", "role": "benchmark-publisher", "sourceUrl": "https://github.com/InternRobotics/REAL"}], "ranking": {"90d": {"confidence": "Medium", "coverage": 0.7, "growth": {"components": {"githubStars": {"normalization": "percentile", "normalized": null, "percentile": null, "value": null}, "hfDatasetDownloads": {"normalization": "percentile", "normalized": null, "percentile": null, "value": null}, "hfPaperUpvotes": {"normalization": "percentile", "normalized": null, "percentile": null, "value": null}}, "confidence": "Low", "coverage": 0.0, "elapsedDays": null, "historyDate": null, "rank": null, "score": null, "windowDays": 90}, "level": {"components": {"githubStars": {"normalization": "log-max", "normalized": 0.5396746658767011, "value": 56}, "hfDatasetDownloads": {"normalization": "log-max", "normalized": null, "value": null}, "hfPaperUpvotes": {"normalization": "log-max", "normalized": 0.0, "value": 0}}, "confidence": "Medium", "coverage": 0.7, "rank": 108, "score": 45}, "method": "attention-ranking-v13", "rank": 108, "score": 45}}, "readiness": "Runnable", "recognitionConfidence": 1.0, "relation": "introduces", "releasedAt": "2026-07-15", "source": {"authors": ["Boyu Mi", "Mengchen Ma", "Yifei Yao", "Xing Gao", "Junting Chen", "Yangzi Li", "Zihou Zhu", "Guohao Li", "Zhenfei Yin", "Tai Wang", "Yao Mu", "Jiangmiao Pang", "Hanqing Wang"], "categories": ["cs.CV", "cs.RO"], "id": "2607.13653", "title": "Exploratory, Communicative, and Deployable: Vision-Driven Embodied Agents for Open-World Mobile Manipulation", "type": "arxiv", "url": "https://arxiv.org/abs/2607.13653"}, "sourceUpdatedAt": "2026-07-27T04:02:35Z", "topics": ["Robotics"], "venueAttempts": [{"decisionRaw": "Accepted to ECCV 2026. 57 pages. Code available at https://github.com/InternRobotics/REAL", "evidence": [{"level": "author-claim", "observedAt": "2026-10-14T01:46:58.058199Z", "rawValue": "Accepted to ECCV 2026. 57 pages. Code available at https://github.com/InternRobotics/REAL", "sourceType": "arxiv-comments", "sourceUrl": "https://arxiv.org/abs/2607.13653"}], "reviewStatus": "accepted", "venueName": "ECCV 2026"}], "whyItMatters": "The benchmark targets the gap between simulation and real-world deployment for embodied agents, offering a repeatable evaluation for long-horizon tasks requiring visual grounding and interactive intent disambiguation. It enables systematic comparison of agent frameworks and informs progress toward practical deployment."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
