{"record": {"aliases": [], "artifacts": [{"id": "arxiv:2608.30536", "kind": "paper", "url": "https://arxiv.org/abs/2608.30536"}, {"id": "gh:nubot-nudt/behavior-skill", "kind": "repo", "url": "https://github.com/nubot-nudt/Behavior-Skill"}], "authors": ["Chunyun Ma", "Lun Luo", "Xingjian Luo", "Xiexing Feng", "Hang Zhang", "Wei Liu", "Feng Qiao", "Yaonan Wang", "Huimin Lu", "Xieyuanli Chen"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2608.30536", "source_url": "https://arxiv.org/abs/2608.30536"}, "categories": ["Multimodal", "Robotics & Embodied AI"], "description": {"en": "Evaluates vision-language-action policies on 50 long-horizon mobile manipulation tasks via 235,492 skill instances, with trajectory-level and skill-level metrics."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2608.30536", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2608.30536", "title": "Behavior-Skill"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "key": "claire-radar:2608.30536", "modality": null, "name": "Behavior-Skill", "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": "true", "origin_record_id": "bm_behavior-skill_187dc006", "origin_source": "arxiv", "recognition_confidence": "1.0", "record_sha256": "f9ebfacb731e87824a73bc4439760d08207b736a4e5942907b98aaa3fb462d3d", "relation": "introduces", "review_model": "deepseek-v4-pro", "review_state": "ai-reviewed", "reviewed_at": "2026-09-02T05:23:11.715633Z", "source_url": "https://arxiv.org/abs/2608.30536"}, "publisher": null, "released": "2026-08-31", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2608.30536", "source_url": "https://arxiv.org/abs/2608.30536"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2608-30536", "source": "claire_radar", "source_benchmark_id": "2608.30536", "source_metadata": {"claire_radar": {"adoption30d": null, "aliases": [], "annotation": "Mixed", "applicationDomains": ["Robotics & Embodied AI"], "area": "Multimodal", "attention": {"asOf": "2026-09-02", "attemptedAt": "2026-09-02T05:12:51.621571Z", "githubRepo": "https://github.com/nubot-nudt/Behavior-Skill", "githubScope": "benchmark_repo", "githubStars": null, "hfDailySubmittedAt": null, "hfDataset": null, "hfDatasetDownloads": null, "hfDatasetLikes": null, "hfPaperUpvotes": 0, "hfPaperUrl": "https://huggingface.co/papers/2608.30536", "observedAt": "2026-09-02T05:12:51.621571Z", "signalStatus": {"githubStars": {"checkedAt": "2026-09-02T05:12:51.621571Z", "state": "unavailable"}, "hfDatasetDownloads": {"state": "not_applicable"}, "hfDatasetLikes": {"state": "not_applicable"}, "hfPaperUpvotes": {"observedAt": "2026-09-02T05:12:51.621571Z", "state": "fresh"}}, "status": "fresh"}, "attentionForecast": {"confidence": "Medium", "horizon": "7d", "reason": "The benchmark addresses a clear gap in long-horizon robot learning and includes public artifacts that facilitate adoption.", "score": 60}, "capabilities": [], "confidence": "High", "construction": "Aggregate Existing", "constructionDetail": "Unknown — the indexer does not infer construction details without explicit source evidence.", "copyGeneration": {"generatedAt": "2026-09-02T05:23:11.715633Z", "inputHash": "f68b5ba169608f66a3cddfe8809c2db0eb03cbd34ad6c0819fbf20b1071bca36", "model": "deepseek-v4-pro"}, "curation": {"canonicalNameEvidence": "We present Behavior-Skill, a benchmark that reformulates the learning and evaluation of long-horizon tasks around executable constituent skills.", "canonicalNameSource": "abstract", "decisionReason": "Behavior-Skill is formally named in the abstract, provides a public code repository, and defines skill-level evaluation with restorable states and success conditions.", "model": "deepseek-v4-pro", "reviewedAt": "2026-09-02T05:23:11.715633Z", "state": "ai-reviewed"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "Evaluates vision-language-action policies on 50 long-horizon mobile manipulation tasks via 235,492 skill instances, with trajectory-level and skill-level metrics.", "displayEligible": true, "domainCuration": {"method": "rules-v1", "state": "auto"}, "evaluationMode": "public_reusable", "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 Behavior-Skill, a benchmark that reformulates the learning and evaluation of long-horizon tasks around executable constituent skills."}, "familyId": "bmf_34c155b410d6", "firstSeenAt": "2026-09-01", "heat": null, "id": "bm_behavior-skill_187dc006", "indexedAt": "2026-09-02T05:21:06.631990Z", "industrySectors": ["Robotics"], "links": {"code": "https://github.com/nubot-nudt/Behavior-Skill", "data": null, "paper": "https://arxiv.org/abs/2608.30536", "pdf": "https://arxiv.org/pdf/2608.30536", "project": null, "report": "https://arxiv.org/abs/2608.30536"}, "metrics": [], "motivation": "Reliable execution of long-horizon mobile manipulation tasks remains challenging because overall task success depends on the successful completion of multiple constituent skills.", "name": "Behavior-Skill", "oneLine": "Evaluates vision-language-action policies on 50 long-horizon mobile manipulation tasks via 235,492 skill instances, with trajectory-level and skill-level metrics.", "paperTitle": "Behavior-Skill: A Fine-Grained Benchmark for Evaluating Vision-Language-Action Policies in Long-Horizon Tasks", "primaryDomain": "Robotics & Embodied AI", "publication": {"evidence": null, "evidenceLevel": "unverified", "evidenceUrl": "https://arxiv.org/abs/2608.30536", "source": "arxiv-metadata", "status": "unverified", "venue": null, "verifiedAt": "2026-09-02T05:21:06.631990Z"}, "publications": [], "publishers": [{"name": "Behavior-Skill team (nubot-nudt)", "organizationType": "academic-lab", "role": "benchmark-publisher", "sourceUrl": "https://github.com/nubot-nudt/Behavior-Skill"}], "ranking": {"30d": {"confidence": "Low", "coverage": 0.3, "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": 30}, "level": {"components": {"githubStars": {"normalization": "log-max", "normalized": null, "value": null}, "hfDatasetDownloads": {"normalization": "log-max", "normalized": null, "value": null}, "hfPaperUpvotes": {"normalization": "log-max", "normalized": 0.0, "value": 0}}, "confidence": "Low", "coverage": 0.3, "rank": null, "score": 35}, "method": "attention-ranking-v13", "rank": null, "score": 35}, "90d": {"confidence": "Low", "coverage": 0.15, "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": null, "value": null}, "hfDatasetDownloads": {"normalization": "log-max", "normalized": null, "value": null}, "hfPaperUpvotes": {"normalization": "log-max", "normalized": 0.0, "value": 0}}, "confidence": "Low", "coverage": 0.15, "rank": null, "score": 43}, "method": "attention-ranking-v13", "rank": null, "score": 43}}, "readiness": "Runnable", "recognitionConfidence": 1.0, "relation": "introduces", "releasedAt": "2026-08-31", "source": {"authors": ["Chunyun Ma", "Lun Luo", "Xingjian Luo", "Xiexing Feng", "Hang Zhang", "Wei Liu", "Feng Qiao", "Yaonan Wang", "Huimin Lu", "Xieyuanli Chen"], "categories": ["cs.RO"], "id": "2608.30536", "title": "Behavior-Skill: A Fine-Grained Benchmark for Evaluating Vision-Language-Action Policies in Long-Horizon Tasks", "type": "arxiv", "url": "https://arxiv.org/abs/2608.30536"}, "sourceUpdatedAt": "2026-08-31T10:03:48Z", "topics": ["Multimodal"], "venueAttempts": [], "whyItMatters": "Moves evaluation beyond aggregate task success to expose intermediate skill bottlenecks, enabling targeted analysis and improvement of long-horizon VLA policies."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
