{"record": {"aliases": [], "artifacts": [{"id": "arxiv:2608.13049", "kind": "paper", "url": "https://arxiv.org/abs/2608.13049"}, {"id": "gh:rongdingyi/h2r-bench", "kind": "repo", "url": "https://github.com/Rongdingyi/H2R-Bench"}], "authors": ["Dingyi Rong", "Yue Shi", "Chaofan Ma", "Jiezhang Cao", "Zongrui Wang", "Zeyu Zhang", "Yao Mu", "Guangtao Zhai", "Ning Liu"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2608.13049", "source_url": "https://arxiv.org/abs/2608.13049"}, "categories": ["Robot manipulation", "Robotics", "Robotics & Embodied AI"], "description": {"en": "H2R-Bench evaluates cross-embodiment human-to-robot manipulation video generation. Models convert egocentric human demonstrations into robot manipulation videos under specified target embodiments. Scoring covers five dimensions: goal-state completion, action-event completion, functional contact transfer, embodiment correctness, and general video quality, aggregated into H2RCore."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2608.13049", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2608.13049", "title": "H2R-Bench"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "institutions": ["Shanghai Jiao Tong University", "Shanghai Artificial Intelligence Laboratory"], "institutions_reference": {"basis": "openalex_linked_institutions", "doi": "10.48550/arxiv.2608.13049", "retrieved_at": "2026-11-04", "source_key": "claire-radar:2608.13049", "source_url": "https://openalex.org/W7203441037"}, "key": "claire-radar:2608.13049", "modality": null, "name": "H2R-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_h2r-bench_da8120f6", "origin_source": "arxiv", "recognition_confidence": "1.0", "record_sha256": "aa264302a0d5b340aec28e45641300904a100bae397b947acc9f718e33c4fe60", "relation": "introduces", "review_state": "unreviewed", "source_url": "https://arxiv.org/abs/2608.13049"}, "publisher": null, "released": "2026-08-13", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2608.13049", "source_url": "https://arxiv.org/abs/2608.13049"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2608-13049", "source": "claire_radar", "source_benchmark_id": "2608.13049", "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/Rongdingyi/H2R-Bench", "githubScope": "benchmark_repo", "githubStars": 4, "hfDailySubmittedAt": null, "hfDataset": null, "hfDatasetDownloads": null, "hfDatasetLikes": null, "hfPaperUpvotes": 21, "hfPaperUrl": "https://huggingface.co/papers/2608.13049", "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", "lastSuccessfulDate": "2026-09-30", "state": "stale"}}, "status": "fresh"}, "capabilities": ["Robot manipulation"], "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": "ff0b215ab2661583ad04130b940ce1173170b802082fd527b01890a9a7ab5e0a", "model": "deepseek-v4-flash"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "H2R-Bench evaluates cross-embodiment human-to-robot manipulation video generation. Models convert egocentric human demonstrations into robot manipulation videos under specified target embodiments. Scoring covers five dimensions: goal-state completion, action-event completion, functional contact transfer, embodiment correctness, and general video quality, aggregated into H2RCore.", "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": "Therefore, we introduce H2R-Bench, a benchmark for evaluating cross-embodiment human-to-robot manipulation video generation, where models transform egocentric human demonstrations into robot manipulation videos under specified embodiments."}, "familyId": "bmf_6a015e569fc5", "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/Rongdingyi/H2R-Bench", "url": "https://github.com/Rongdingyi/H2R-Bench"}, {"basis": "existing-metric-association", "linkedUrl": "https://github.com/Rongdingyi/H2R-Bench", "url": "https://huggingface.co/papers/2608.13049"}], "scope": "benchmark_repo", "url": "https://github.com/Rongdingyi/H2R-Bench", "verification": "source-linked"}], "status": "linked"}, "heat": null, "id": "bm_h2r-bench_da8120f6", "indexedAt": "2026-08-19T11:05:13.395391Z", "industrySectors": ["Robotics"], "links": {"code": "https://github.com/Rongdingyi/H2R-Bench", "data": null, "hfPaper": "https://huggingface.co/papers/2608.13049", "paper": "https://arxiv.org/abs/2608.13049", "pdf": "https://arxiv.org/pdf/2608.13049", "project": null, "report": "https://arxiv.org/abs/2608.13049"}, "metrics": [], "motivation": "Large-scale manipulation data is essential for robot learning, yet collecting robot demonstrations remains expensive and difficult to scale.", "name": "H2R-Bench", "oneLine": "H2R-Bench evaluates cross-embodiment human-to-robot manipulation video generation. Models convert egocentric human demonstrations into robot manipulation videos under specified target embodiments. Scoring covers five dimensions: goal-state completion, action-event completion, functional contact transfer, embodiment correctness, and general video quality, aggregated into H2RCore.", "paperTitle": "H2R-Bench: Benchmarking Human-to-Robot Manipulation Video Generation in World Models", "primaryDomain": "Robotics & Embodied AI", "publication": {"evidence": null, "evidenceLevel": "unverified", "evidenceUrl": "https://arxiv.org/abs/2608.13049", "source": "arxiv-metadata", "status": "unverified", "venue": null, "verifiedAt": "2026-10-14T01:46:58.058199Z"}, "publications": [], "publishers": [{"name": "Shanghai Jiao Tong University", "organizationType": "academic-lab", "role": "benchmark-publisher", "sourceUrl": "https://github.com/Rongdingyi/H2R-Bench"}, {"name": "Shanghai Artificial Intelligence Laboratory", "organizationType": "academic-lab", "role": "benchmark-publisher", "sourceUrl": "https://github.com/Rongdingyi/H2R-Bench"}], "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": 0.21483102985913266, "value": 4}, "hfDatasetDownloads": {"normalization": "log-max", "normalized": null, "value": null}, "hfPaperUpvotes": {"normalization": "log-max", "normalized": 0.5461705391971857, "value": 21}}, "confidence": "Medium", "coverage": 0.7, "rank": 188, "score": 35}, "method": "attention-ranking-v13", "rank": 188, "score": 35}}, "readiness": "Runnable", "recognitionConfidence": 1.0, "relation": "introduces", "releasedAt": "2026-08-13", "source": {"authors": ["Dingyi Rong", "Yue Shi", "Chaofan Ma", "Jiezhang Cao", "Zongrui Wang", "Zeyu Zhang", "Yao Mu", "Guangtao Zhai", "Ning Liu"], "categories": ["cs.RO", "cs.CV"], "id": "2608.13049", "title": "H2R-Bench: Benchmarking Human-to-Robot Manipulation Video Generation in World Models", "type": "arxiv", "url": "https://arxiv.org/abs/2608.13049"}, "sourceUpdatedAt": "2026-08-13T10:14:33Z", "topics": ["Robotics"], "venueAttempts": [], "whyItMatters": "Assesses whether video world models can bridge the embodiment gap between human hands and robotic end-effectors, providing a diagnostic for how well generated videos transfer functional interactions and task execution. Results show generic video quality does not correlate with transfer validity, aiding model selection for robot learning from human video."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
