{"record": {"aliases": [], "artifacts": [{"id": "arxiv:2607.00491", "kind": "paper", "url": "https://arxiv.org/abs/2607.00491"}, {"id": "hf:zodaofficial/mindedit-bench", "kind": "dataset", "url": "https://huggingface.co/datasets/ZODAOfficial/MindEdit-Bench"}], "authors": ["Leyuan Yu", "Xiao Tang", "Minghao Liu", "Xinyuan Li", "Xiaokai Bai", "Sheng Zhou", "Qunshu Lin", "Weihao Xuan", "Naoto Yokoya"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2607.00491", "source_url": "https://arxiv.org/abs/2607.00491"}, "categories": ["Geometric reasoning", "Mobile & Personal Computing", "Reasoning", "Vision & 3D"], "description": {"en": "MindEdit-Bench evaluates counterfactual spatial reasoning in VLMs with 1,003 multiple-choice questions from private indoor scenes, covering six task types."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2607.00491", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2607.00491", "title": "MindEdit-Bench"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "institutions": ["Zhejiang University", "Tongji University", "The University of Tokyo"], "institutions_reference": {"basis": "openalex_linked_institutions", "doi": "10.48550/arxiv.2607.00491", "retrieved_at": "2026-11-04", "source_key": "claire-radar:2607.00491", "source_url": "https://openalex.org/W7167020377"}, "key": "claire-radar:2607.00491", "modality": null, "name": "MindEdit-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_mindedit-bench_02b6e953", "origin_source": "arxiv", "recognition_confidence": "1.0", "record_sha256": "69089ab49901772f027fc5b898d3048bf66be5d5a274a9fd157921e1526fb8bd", "relation": "introduces", "review_state": "unreviewed", "source_url": "https://arxiv.org/abs/2607.00491"}, "publisher": null, "released": "2026-07-01", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2607.00491", "source_url": "https://arxiv.org/abs/2607.00491"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2607-00491", "source": "claire_radar", "source_benchmark_id": "2607.00491", "source_metadata": {"claire_radar": {"adoption30d": null, "aliases": [], "annotation": "Unknown", "applicationDomains": ["Mobile & Personal Computing"], "area": "Vision & 3D", "attention": {"asOf": "2026-10-05", "attemptedAt": "2026-10-04T15:31:48.030916Z", "githubRepo": null, "githubScope": null, "githubStars": null, "hfDailySubmittedAt": null, "hfDataset": "ZODAOfficial/MindEdit-Bench", "hfDatasetDownloads": 135, "hfDatasetLikes": 1, "hfPaperUpvotes": 0, "hfPaperUrl": "https://huggingface.co/papers/2607.00491", "observedAt": "2026-10-04T03:02:42.754940Z", "signalStatus": {"githubStars": {"state": "not_applicable"}, "hfDatasetDownloads": {"checkedAt": "2026-10-04T15:31:48.030916Z", "lastSuccessfulDate": "2026-10-04", "state": "stale"}, "hfDatasetLikes": {"checkedAt": "2026-10-04T15:31:48.030916Z", "lastSuccessfulDate": "2026-10-04", "state": "stale"}, "hfPaperUpvotes": {"checkedAt": "2026-10-04T15:31:48.030916Z", "lastSuccessfulDate": "2026-10-04", "state": "stale"}}, "status": "stale"}, "capabilities": ["Reasoning", "Geometric 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": "d54d6d1a7d263f5a4b25aa43486eee2c5b6fccc75f219a1ad4181db3e06a7e67", "model": "deepseek-v4-flash"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "MindEdit-Bench evaluates counterfactual spatial reasoning in VLMs with 1,003 multiple-choice questions from private indoor scenes, covering six task types.", "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", "public artifact URL"], "snippet": "We introduce MindEdit-Bench, a benchmark of six spatial reasoning tasks built from three-photo smartphone triplets of newly captured indoor scenes via an automatic in-the-wild 3D scene-graph extraction pipeline."}, "familyId": "bmf_ebce6cd9eb55", "firstSeenAt": "2026-08-19", "githubIndex": {"candidateCount": 0, "indexedAt": "2026-09-17T04:40:34.111665+00:00", "repositories": [], "status": "not-found"}, "heat": null, "id": "bm_mindedit-bench_02b6e953", "indexedAt": "2026-08-19T11:05:13.395391Z", "industrySectors": ["Consumer Technology"], "links": {"code": null, "data": "https://huggingface.co/datasets/ZODAOfficial/MindEdit-Bench", "hfPaper": "https://huggingface.co/papers/2607.00491", "paper": "https://arxiv.org/abs/2607.00491", "pdf": "https://arxiv.org/pdf/2607.00491", "project": null, "report": "https://arxiv.org/abs/2607.00491"}, "metrics": [], "motivation": "Benchmarks for vision-language models (VLMs) mostly test observational spatial reasoning: models describe relations already visible in the input.", "name": "MindEdit-Bench", "oneLine": "MindEdit-Bench evaluates counterfactual spatial reasoning in VLMs with 1,003 multiple-choice questions from private indoor scenes, covering six task types.", "paperTitle": "MindEdit-Bench: Benchmarking Object-Level Counterfactual Spatial Reasoning in VLMs from In-the-Wild Photos", "primaryDomain": "Mobile & Personal Computing", "publication": {"evidence": null, "evidenceLevel": "unverified", "evidenceUrl": "https://arxiv.org/abs/2607.00491", "source": "arxiv-metadata", "status": "unverified", "venue": null, "verifiedAt": "2026-10-14T01:46:58.058199Z"}, "publications": [], "publishers": [{"name": "ZODAOfficial", "organizationType": "community", "role": "benchmark-publisher", "sourceUrl": "https://huggingface.co/datasets/ZODAOfficial/MindEdit-Bench"}], "ranking": {}, "readiness": "Inspectable", "recognitionConfidence": 1.0, "relation": "introduces", "releasedAt": "2026-07-01", "source": {"authors": ["Leyuan Yu", "Xiao Tang", "Minghao Liu", "Xinyuan Li", "Xiaokai Bai", "Sheng Zhou", "Qunshu Lin", "Weihao Xuan", "Naoto Yokoya"], "categories": ["cs.CV", "cs.AI", "cs.CL"], "id": "2607.00491", "title": "MindEdit-Bench: Benchmarking Object-Level Counterfactual Spatial Reasoning in VLMs from In-the-Wild Photos", "type": "arxiv", "url": "https://arxiv.org/abs/2607.00491"}, "sourceUpdatedAt": "2026-07-01T06:19:54Z", "topics": ["Reasoning"], "venueAttempts": [], "whyItMatters": "Tests whether VLMs can reason about hypothetical object manipulations, a capability not covered by existing benchmarks, with human-verified answers."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
