{"record": {"aliases": ["Metric-Bench"], "artifacts": [{"id": "hf:yulingxi/metric-bench", "kind": "dataset", "url": "https://huggingface.co/datasets/yulingxi/Metric-Bench"}], "categories": ["General AI", "Language & Knowledge", "cs.AI"], "description": {"en": "A reconstructed test release for metric spatial understanding over indoor RGB images with supplied anchor measurements. Each example pairs one image and one English question with a numeric JSON reference answer, covering width, length, height, dimension, center, depth, elevation, distance and box measurements."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://huggingface.co/datasets/yulingxi/Metric-Bench", "source": "claire_radar", "source_url": "https://huggingface.co/datasets/yulingxi/Metric-Bench", "title": "Metric-Bench Test"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "key": "claire-radar:huggingface:yulingxi/metric-bench", "modality": null, "name": "Metric-Bench Test", "openness": {"code_license": null, "data_license": null, "evidence": [], "status": "unknown"}, "provenance": {"admission_policy_version": "2026-09-09.1", "confidence": "Medium", "crawl_bundle": "claire_radar_2026-10-14", "crawled_at": "2026-10-14T09:44:51Z", "data_status": "primary-source-candidate", "display_eligible": "false", "origin_record_id": "bm_metric-bench_e77780e5", "origin_source": "huggingface", "recognition_confidence": "0.4", "record_sha256": "6551f2b880eb1a148b8a2de5140725ac037d912a20b770cd7b573de380eac9e0", "relation": "unclear", "review_model": "deepseek-v4-pro", "review_state": "ai-name-audit-deferred", "reviewed_at": "2026-09-14T06:44:31.931424Z", "source_url": "https://huggingface.co/datasets/yulingxi/Metric-Bench"}, "publisher": null, "released": "2026-09-11", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:huggingface:yulingxi/metric-bench", "source_url": "https://huggingface.co/datasets/yulingxi/Metric-Bench"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-huggingface-yulingxi-metric-bench", "source": "claire_radar", "source_benchmark_id": "huggingface:yulingxi/metric-bench", "source_metadata": {"claire_radar": {"adoption30d": null, "aliases": ["Metric-Bench"], "annotation": "Unknown", "applicationDomains": ["General AI"], "area": "Language & Knowledge", "attention": {"asOf": "2026-09-13", "attemptedAt": "2026-09-13T01:36:51.692703Z", "githubRepo": null, "githubScope": null, "githubStars": null, "hfDailySubmittedAt": null, "hfDataset": "yulingxi/Metric-Bench", "hfDatasetDownloads": 23, "hfDatasetLikes": 0, "hfPaperUpvotes": null, "hfPaperUrl": null, "observedAt": "2026-09-13T01:36:51.692703Z", "signalStatus": {"githubStars": {"state": "not_applicable"}, "hfDatasetDownloads": {"observedAt": "2026-09-13T01:36:51.692703Z", "state": "fresh"}, "hfDatasetLikes": {"observedAt": "2026-09-13T01:36:51.692703Z", "state": "fresh"}, "hfPaperUpvotes": {"state": "not_applicable"}}, "status": "fresh"}, "attentionForecast": {"confidence": "Low", "horizon": "7d", "reason": "Attention is likely bounded by its niche spatial-reasoning topic, its reconstructed test-set scope, and the absence of a separate project page or paper linkage.", "score": 18}, "capabilities": [], "confidence": "Medium", "construction": "Unknown", "constructionDetail": "Unknown — the indexer does not infer construction details without explicit source evidence.", "copyGeneration": {"generatedAt": "2026-09-14T10:07:09.772218Z", "inputHash": "b00135a02ad2085e0f946005176e1f202ab39e31f9a2ed55852c3b5bdb3a885b", "model": "deepseek-v4-flash"}, "curation": {"admissionPolicyVersion": "2026-09-09.1", "canonicalNameEvidence": "pretty_name: Metric-Bench Test", "canonicalNameSource": "official_readme", "decisionReason": "The Hugging Face dataset contains a test split with clear task definition and reusable data, though lack of an official scoring implementation makes the scoring contract incomplete.", "evaluationEvidence": {"reuse": "This release provides reference answers but does not ship an official scoring implementation or specify a paper-equivalent acceptance threshold.", "scoring": "Ground truth uses a JSON string so scalar and vector answers share one stable column type.", "task": "Metric-Bench Test set evaluates metric spatial understanding from indoor RGB images and explicit anchor measurements."}, "model": "deepseek-v4-pro", "reviewedAt": "2026-09-14T06:44:31.931424Z", "state": "ai-name-audit-deferred"}, "dataStatus": "primary-source-candidate", "demo": false, "description": "A reconstructed test release for metric spatial understanding over indoor RGB images with supplied anchor measurements. Each example pairs one image and one English question with a numeric JSON reference answer, covering width, length, height, dimension, center, depth, elevation, distance and box measurements.", "discoverySources": [{"id": "huggingface:yulingxi/metric-bench", "type": "huggingface", "url": "https://huggingface.co/datasets/yulingxi/Metric-Bench"}], "displayEligible": false, "domainCuration": {"method": "rules-v1", "state": "auto"}, "evaluationMode": "public_reusable", "evidence": {"reasonCodes": ["discovered via huggingface", "benchmark term in abstract", "evaluation protocol evidence", "public artifact URL", "no explicit benchmark release evidence"], "snippet": "Metric-Bench Test Metric-Bench Test set evaluates metric spatial understanding from indoor RGB images and explicit anchor measurements."}, "familyId": "bmf_c327ee6b71e1", "firstSeenAt": "2026-09-12", "heat": null, "id": "bm_metric-bench_e77780e5", "indexedAt": "2026-09-12T00:52:13.701164Z", "industrySectors": [], "links": {"data": "https://huggingface.co/datasets/yulingxi/Metric-Bench", "paper": "https://huggingface.co/datasets/yulingxi/Metric-Bench", "report": "https://huggingface.co/datasets/yulingxi/Metric-Bench"}, "metrics": [], "motivation": "Metric-Bench Test Metric-Bench Test set evaluates metric spatial understanding from indoor RGB images and explicit anchor measurements.", "name": "Metric-Bench Test", "oneLine": "A reconstructed test release for metric spatial understanding over indoor RGB images with supplied anchor measurements. Each example pairs one image and one English question with a numeric JSON reference answer, covering width, length, height, dimension, center, depth, elevation, distance and box measurements.", "paperTitle": "Metric-Bench", "primaryDomain": "General AI", "publication": {"evidence": null, "evidenceLevel": "unverified", "evidenceUrl": "https://huggingface.co/datasets/yulingxi/Metric-Bench", "source": "arxiv-metadata", "status": "unverified", "venue": null, "verifiedAt": "2026-09-12T00:52:13.701164Z"}, "publications": [], "ranking": {"30d": {"confidence": "Low", "coverage": 0.15, "datasetDownloadRank": 42, "datasetRankPopulation": 42, "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": 0.3250988395760043, "value": 23}, "hfPaperUpvotes": {"normalization": "log-max", "normalized": null, "value": null}}, "confidence": "Low", "coverage": 0.15, "rank": 43, "score": 47}, "method": "attention-ranking-v13", "rank": 43, "score": 47}, "90d": {"confidence": "Low", "coverage": 0.3, "datasetDownloadRank": 74, "datasetRankPopulation": 76, "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": 0.2629018976473374, "value": 23}, "hfPaperUpvotes": {"normalization": "log-max", "normalized": null, "value": null}}, "confidence": "Low", "coverage": 0.3, "rank": 129, "score": 43}, "method": "attention-ranking-v13", "rank": 129, "score": 43}}, "readiness": "Inspectable", "recognitionConfidence": 0.4, "relation": "unclear", "releasedAt": "2026-09-11", "source": {"authors": ["yulingxi"], "categories": [], "id": "huggingface:yulingxi/metric-bench", "publicSignals": {"hfDatasetDownloads": 0, "hfDatasetLikes": 0}, "title": "Metric-Bench", "type": "huggingface", "url": "https://huggingface.co/datasets/yulingxi/Metric-Bench"}, "sourceUpdatedAt": "2026-09-11T10:28:44.000Z", "topics": ["cs.AI"], "venueAttempts": [], "whyItMatters": "It offers a held-out resource for measuring visual metric estimation with explicit anchors rather than object detection or unassisted estimation. Reporting must state revision, prompt, decoding settings, invalid-output handling and numeric tolerance, because no official scoring implementation or paper-equivalent threshold ships with the release."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
