{"record": {"aliases": [], "artifacts": [{"id": "arxiv:2607.24957", "kind": "paper", "url": "https://arxiv.org/abs/2607.24957"}], "authors": ["Zichao Lin", "Yifeng Xie", "Bowen Qu", "Haiming Wang", "Jia Li", "Haoning Wu", "Yuhao Dong", "Zuhao Yang", "Jinguo Zhu", "Haoyu Lu", "Zijia Zhao", "Tongtian Yue", "Zhangyang Qi", "Junwei Yang", "Mengfan Dong", "Peizhou Cao", "Chenzhuang Du", "Zaida Zhou", "Haotian Yao", "Hao Yang", "Hongcheng Gao", "Lin Sui", "Weihong Li", "Xinxing Zu", "Jia Chen", "Yao Wang", "Xiaoxue Wu", "Yalin Wang", "Y. Charles", "Yiping Bao", "Yangyang Liu", "Zhiqi Huang", "Xinyu Zhou"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2607.24957", "source_url": "https://arxiv.org/abs/2607.24957"}, "categories": ["General AI", "Multimodal"], "description": {"en": "PerceptionBench evaluates atomic visual perception in MLLMs with 3,000 verified questions isolating ten perceptual capabilities, based on an error taxonomy from 42 benchmarks."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2607.24957", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2607.24957", "title": "PerceptionBench"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "key": "claire-radar:2607.24957", "modality": null, "name": "PerceptionBench", "openness": {"code_license": null, "data_license": null, "evidence": [], "status": "unknown"}, "provenance": {"confidence": "Medium", "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_perceptionbench_27ae03ae", "origin_source": "arxiv", "recognition_confidence": "0.85", "record_sha256": "ca160038af212909f8300ebb436733f091814379b228f7b6480d028afa51a48f", "relation": "introduces", "review_state": "unreviewed", "source_url": "https://arxiv.org/abs/2607.24957"}, "publisher": null, "released": "2026-07-27", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2607.24957", "source_url": "https://arxiv.org/abs/2607.24957"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2607-24957", "source": "claire_radar", "source_benchmark_id": "2607.24957", "source_metadata": {"claire_radar": {"adoption30d": null, "aliases": [], "annotation": "Unknown", "applicationDomains": ["General AI"], "area": "Multimodal", "attention": {"asOf": "2026-10-05", "attemptedAt": "2026-10-04T15:31:48.030916Z", "githubRepo": null, "githubScope": null, "githubStars": null, "hfDailySubmittedAt": "2026-07-29T00:00:00.000Z", "hfDataset": null, "hfDatasetDownloads": null, "hfDatasetLikes": null, "hfPaperUpvotes": 15, "hfPaperUrl": "https://huggingface.co/papers/2607.24957", "observedAt": "2026-10-04T15:31:48.030916Z", "signalStatus": {"githubStars": {"state": "not_applicable"}, "hfDatasetDownloads": {"state": "not_applicable"}, "hfDatasetLikes": {"state": "not_applicable"}, "hfPaperUpvotes": {"observedAt": "2026-10-04T15:31:48.030916Z", "state": "fresh"}}, "status": "fresh"}, "capabilities": [], "confidence": "Medium", "construction": "Unknown", "constructionDetail": "Unknown — the indexer does not infer construction details without explicit source evidence.", "copyGeneration": {"generatedAt": "2026-08-20T15:22:10.044731Z", "inputHash": "1dbc8e54ea45881534d104682db2e2e19d0ded65ba923bbca03fd4741fd614d7", "model": "deepseek-v4-flash"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "PerceptionBench evaluates atomic visual perception in MLLMs with 3,000 verified questions isolating ten perceptual capabilities, based on an error taxonomy from 42 benchmarks.", "domainCuration": {"method": "rules-v1", "state": "auto"}, "evidence": {"reasonCodes": ["coined title prefix ending in Bench or Benchmark", "benchmark release stated in one sentence"], "snippet": "We introduce PerceptionBench, a benchmark specifically designed to evaluate the atomic visual perception capabilities of Multimodal Large Language Models (MLLMs)."}, "familyId": "bmf_885ad8ebb2ad", "firstSeenAt": "2026-08-19", "githubIndex": {"candidateCount": 0, "indexedAt": "2026-09-17T04:40:34.111665+00:00", "repositories": [], "status": "not-found"}, "heat": null, "id": "bm_perceptionbench_27ae03ae", "indexedAt": "2026-08-19T11:05:13.395391Z", "industrySectors": [], "links": {"code": null, "data": null, "hfPaper": "https://huggingface.co/papers/2607.24957", "paper": "https://arxiv.org/abs/2607.24957", "pdf": "https://arxiv.org/pdf/2607.24957", "project": null, "report": "https://arxiv.org/abs/2607.24957"}, "metrics": [], "motivation": "We introduce PerceptionBench, a benchmark specifically designed to evaluate the atomic visual perception capabilities of Multimodal Large Language Models (MLLMs).", "name": "PerceptionBench", "oneLine": "PerceptionBench evaluates atomic visual perception in MLLMs with 3,000 verified questions isolating ten perceptual capabilities, based on an error taxonomy from 42 benchmarks.", "paperTitle": "PerceptionBench: Evaluating Atomic Visual Perception in Multimodal Large Language Models", "primaryDomain": "General AI", "publication": {"evidence": null, "evidenceLevel": "unverified", "evidenceUrl": "https://arxiv.org/abs/2607.24957", "source": "arxiv-metadata", "status": "unverified", "venue": null, "verifiedAt": "2026-10-14T01:46:58.058199Z"}, "publications": [], "ranking": {"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": 0.48990148153820845, "value": 15}}, "confidence": "Low", "coverage": 0.15, "rank": null, "score": 50}, "method": "attention-ranking-v13", "rank": null, "score": 50}}, "readiness": "Paper only", "recognitionConfidence": 0.85, "relation": "introduces", "releasedAt": "2026-07-27", "source": {"authors": ["Zichao Lin", "Yifeng Xie", "Bowen Qu", "Haiming Wang", "Jia Li", "Haoning Wu", "Yuhao Dong", "Zuhao Yang", "Jinguo Zhu", "Haoyu Lu", "Zijia Zhao", "Tongtian Yue", "Zhangyang Qi", "Junwei Yang", "Mengfan Dong", "Peizhou Cao", "Chenzhuang Du", "Zaida Zhou", "Haotian Yao", "Hao Yang", "Hongcheng Gao", "Lin Sui", "Weihong Li", "Xinxing Zu", "Jia Chen", "Yao Wang", "Xiaoxue Wu", "Yalin Wang", "Y. Charles", "Yiping Bao", "Yangyang Liu", "Zhiqi Huang", "Xinyu Zhou"], "categories": ["cs.CV"], "id": "2607.24957", "title": "PerceptionBench: Evaluating Atomic Visual Perception in Multimodal Large Language Models", "type": "arxiv", "url": "https://arxiv.org/abs/2607.24957"}, "sourceUpdatedAt": "2026-07-27T18:04:54Z", "topics": ["Multimodal"], "venueAttempts": [], "whyItMatters": "Addresses the need for a capability-level standard to diagnose visual perception boundaries, showing that current MLLMs remain below 60% accuracy."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
