{"record": {"aliases": [], "artifacts": [{"id": "arxiv:2606.17639", "kind": "paper", "url": "https://arxiv.org/abs/2606.17639"}, {"id": "gh:lunaproject22/erqa-plus", "kind": "repo", "url": "https://github.com/LUNAProject22/erqa-plus"}, {"id": "hf:huggingdas/erqa-plus", "kind": "dataset", "url": "https://huggingface.co/datasets/huggingdas/erqa-plus"}], "authors": ["Hong Yang", "Basura Fernando"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2606.17639", "source_url": "https://arxiv.org/abs/2606.17639"}, "categories": ["Reasoning", "Robotics & Embodied AI"], "description": {"en": "ERQA-Plus evaluates embodied reasoning in AI systems with 1,766 question-answer instances grounded in 711 robot-centric images, organized by a taxonomy covering perceptual, action-centric, social-interaction, navigation-environmental, and commonsense reasoning. Scoring uses overall accuracy and SBERT similarity."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2606.17639", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2606.17639", "title": "ERQA-Plus"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "institutions": ["Agency for Science, Technology and Research", "Nanyang Technological University"], "institutions_reference": {"basis": "openalex_linked_institutions", "doi": "10.48550/arxiv.2606.17639", "retrieved_at": "2026-11-04", "source_key": "claire-radar:2606.17639", "source_url": "https://openalex.org/W7165025162"}, "key": "claire-radar:2606.17639", "modality": null, "name": "ERQA-Plus", "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_erqa-plus_148e4067", "origin_source": "arxiv", "recognition_confidence": "1.0", "record_sha256": "5fc543ed82aef42ce8c80afa97b0bd1035b4df13b145ef0ce90aa2cfd758b652", "relation": "introduces", "review_state": "unreviewed", "source_url": "https://arxiv.org/abs/2606.17639"}, "publisher": null, "released": "2026-06-16", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2606.17639", "source_url": "https://arxiv.org/abs/2606.17639"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2606-17639", "source": "claire_radar", "source_benchmark_id": "2606.17639", "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/LUNAProject22/erqa-plus", "githubScope": "benchmark_repo", "githubStars": null, "hfDailySubmittedAt": null, "hfDataset": "huggingdas/erqa-plus", "hfDatasetDownloads": 348, "hfDatasetLikes": 0, "hfPaperUpvotes": null, "hfPaperUrl": "https://huggingface.co/papers/2606.17639", "observedAt": "2026-10-04T03:02:42.754940Z", "signalStatus": {"githubStars": {"checkedAt": "2026-10-04T15:31:48.030916Z", "state": "unavailable"}, "hfDatasetDownloads": {"checkedAt": "2026-10-04T15:31:48.030916Z", "lastSuccessfulDate": "2026-10-01", "state": "stale"}, "hfDatasetLikes": {"checkedAt": "2026-10-04T15:31:48.030916Z", "lastSuccessfulDate": "2026-10-01", "state": "stale"}, "hfPaperUpvotes": {"checkedAt": "2026-10-04T15:31:48.030916Z", "state": "unavailable"}}, "status": "stale"}, "capabilities": ["Reasoning"], "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": "7791fd2963a0f20915e4425ee5ea4b82d7d2e57de5e196da5494206fd71fdd8e", "model": "deepseek-v4-flash"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "ERQA-Plus evaluates embodied reasoning in AI systems with 1,766 question-answer instances grounded in 711 robot-centric images, organized by a taxonomy covering perceptual, action-centric, social-interaction, navigation-environmental, and commonsense reasoning. Scoring uses overall accuracy and SBERT similarity.", "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", "public artifact URL"], "snippet": "We present ERQA-Plus, a diagnostic benchmark for reasoning in embodied AI."}, "familyId": "bmf_43898951b3ad", "firstSeenAt": "2026-08-19", "githubIndex": {"candidateCount": 2, "indexedAt": "2026-09-17T04:40:34.111665+00:00", "repositories": [], "status": "candidates"}, "heat": null, "id": "bm_erqa-plus_148e4067", "indexedAt": "2026-08-19T11:05:13.395391Z", "industrySectors": ["Robotics"], "links": {"code": "https://github.com/LUNAProject22/erqa-plus", "data": "https://huggingface.co/datasets/huggingdas/erqa-plus", "hfPaper": "https://huggingface.co/papers/2606.17639", "paper": "https://arxiv.org/abs/2606.17639", "pdf": "https://arxiv.org/pdf/2606.17639", "project": null, "report": "https://arxiv.org/abs/2606.17639"}, "metrics": [], "motivation": "Generalist embodied agents require more than object recognition: they must reason about spatial relations, actions, procedures, human intentions, environmental constraints, and commonsense consequences from situated visual observations.", "name": "ERQA-Plus", "oneLine": "ERQA-Plus evaluates embodied reasoning in AI systems with 1,766 question-answer instances grounded in 711 robot-centric images, organized by a taxonomy covering perceptual, action-centric, social-interaction, navigation-environmental, and commonsense reasoning. Scoring uses overall accuracy and SBERT similarity.", "paperTitle": "ERQA-Plus: A Diagnostic Benchmark for Reasoning in Embodied AI", "primaryDomain": "Robotics & Embodied AI", "publication": {"evidence": null, "evidenceLevel": "unverified", "evidenceUrl": "https://arxiv.org/abs/2606.17639", "source": "arxiv-metadata", "status": "unverified", "venue": null, "verifiedAt": "2026-10-14T01:46:58.058199Z"}, "publications": [], "publishers": [{"name": "LUNAProject22", "organizationType": "community", "role": "benchmark-publisher", "sourceUrl": "https://github.com/LUNAProject22/erqa-plus"}], "ranking": {}, "readiness": "Runnable", "recognitionConfidence": 1.0, "relation": "introduces", "releasedAt": "2026-06-16", "source": {"authors": ["Hong Yang and Basura Fernando"], "categories": ["cs.RO", "cs.CV"], "id": "2606.17639", "title": "ERQA-Plus: A Diagnostic Benchmark for Reasoning in Embodied AI", "type": "arxiv", "url": "https://arxiv.org/abs/2606.17639"}, "sourceUpdatedAt": "2026-06-17T07:18:46Z", "topics": ["Reasoning"], "venueAttempts": [], "whyItMatters": "Existing visual and embodied QA benchmarks often lack control over reasoning dependencies, making it hard to distinguish genuine embodied reasoning from shortcut-driven pattern matching. ERQA-Plus provides a fine-grained diagnostic to identify strengths and weaknesses in specific reasoning categories, informing model development and deployment decisions."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
