{"record": {"aliases": [], "artifacts": [{"id": "arxiv:2608.09281", "kind": "paper", "url": "https://arxiv.org/abs/2608.09281"}, {"kind": "website", "url": "https://dcx-swjtu.github.io/MMArch/"}], "authors": ["Chenxu Du", "Kang An", "Tengyue Wang", "Zhongyu Yang", "Xinqi Yang", "Yuanchi Zhu", "Hebao Zhu", "Ziliang Wang", "Faqiang Qian", "Yunli Yang", "Qibing Ren"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2608.09281", "source_url": "https://arxiv.org/abs/2608.09281"}, "categories": ["General AI", "Multimodal", "Reasoning"], "description": {"en": "MMArch is a benchmark for multimodal reasoning in architecture and civil engineering, spanning ten subdomains and built from figures in peer-reviewed papers. It contains 1,212 short-answer items requiring perception, principle identification, and application."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2608.09281", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2608.09281", "title": "MMArch"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "institutions": ["Southwest Jiaotong University", "South China University of Technology", "East China Normal University", "Chinese Academy of Sciences", "ShanghaiTech University", "Institute of Automation", "Chongqing University", "Sensimetrics Corporation", "Shanghai Jiao Tong University"], "institutions_reference": {"basis": "openalex_linked_institutions", "doi": "10.48550/arxiv.2608.09281", "retrieved_at": "2026-11-04", "source_key": "claire-radar:2608.09281", "source_url": "https://openalex.org/W7202206228"}, "key": "claire-radar:2608.09281", "modality": null, "name": "MMArch", "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_mmarch_3c23535d", "origin_source": "arxiv", "recognition_confidence": "1.0", "record_sha256": "035df38c1050247a88b665548a25329b12ae784c14f0d38590573df8983b881d", "relation": "introduces", "review_state": "unreviewed", "source_url": "https://arxiv.org/abs/2608.09281"}, "publisher": null, "released": "2026-08-10", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2608.09281", "source_url": "https://arxiv.org/abs/2608.09281"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2608-09281", "source": "claire_radar", "source_benchmark_id": "2608.09281", "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": null, "hfDataset": null, "hfDatasetDownloads": null, "hfDatasetLikes": null, "hfPaperUpvotes": null, "hfPaperUrl": "https://huggingface.co/papers/2608.09281", "observedAt": "2026-10-04T03:02:42.754940Z", "signalStatus": {"githubStars": {"state": "not_applicable"}, "hfDatasetDownloads": {"state": "not_applicable"}, "hfDatasetLikes": {"state": "not_applicable"}, "hfPaperUpvotes": {"checkedAt": "2026-10-04T15:31:48.030916Z", "state": "unavailable"}}, "status": "unavailable"}, "capabilities": ["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": "1791fc684d00be2eaa297e2ddec1284cf0fe1b3277486b536ab94ee02070430f", "model": "deepseek-v4-flash"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "MMArch is a benchmark for multimodal reasoning in architecture and civil engineering, spanning ten subdomains and built from figures in peer-reviewed papers. It contains 1,212 short-answer items requiring perception, principle identification, and application.", "domainCuration": {"method": "rules-v1", "state": "auto"}, "evidence": {"reasonCodes": ["exact coined title identity tied to benchmark evidence", "benchmark release stated in one sentence", "public artifact URL"], "snippet": "We introduce MMArch, a benchmark for architecture and civil engineering spanning ten subdomains and built entirely from figures in peer-reviewed papers."}, "familyId": "bmf_cde8c1453aed", "firstSeenAt": "2026-08-19", "githubIndex": {"candidateCount": 0, "indexedAt": "2026-09-17T04:40:34.111665+00:00", "repositories": [], "status": "not-found"}, "heat": null, "id": "bm_mmarch_3c23535d", "indexedAt": "2026-08-19T11:05:13.395391Z", "industrySectors": [], "links": {"code": null, "data": null, "hfPaper": "https://huggingface.co/papers/2608.09281", "paper": "https://arxiv.org/abs/2608.09281", "pdf": "https://arxiv.org/pdf/2608.09281", "project": "https://dcx-swjtu.github.io/MMArch/", "report": "https://arxiv.org/abs/2608.09281"}, "metrics": [], "motivation": "Multimodal large language models (MLLMs) perform strongly on engineering imagery, yet existing benchmarks mostly test drawing recognition, information extraction, or compliance checking, leaving open whether models can combine distributed visual evidence with engineering principles to reach a conclusion.", "name": "MMArch", "oneLine": "MMArch is a benchmark for multimodal reasoning in architecture and civil engineering, spanning ten subdomains and built from figures in peer-reviewed papers. It contains 1,212 short-answer items requiring perception, principle identification, and application.", "paperTitle": "MMArch: Benchmarking Multimodal Reasoning Grounded in Architectural Evidence", "primaryDomain": "General AI", "publication": {"evidence": null, "evidenceLevel": "unverified", "evidenceUrl": "https://arxiv.org/abs/2608.09281", "source": "arxiv-metadata", "status": "unverified", "venue": null, "verifiedAt": "2026-10-14T01:46:58.058199Z"}, "publications": [], "publishers": [{"name": "dcx-swjtu", "organizationType": "academic-lab", "role": "benchmark-publisher", "sourceUrl": "https://dcx-swjtu.github.io/MMArch/"}], "ranking": {"90d": {"confidence": "Low", "coverage": 0.0, "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": null, "value": null}}, "confidence": "Low", "coverage": 0.0, "rank": null, "score": null}, "method": "attention-ranking-v13", "rank": null, "score": null}}, "readiness": "Inspectable", "recognitionConfidence": 1.0, "relation": "introduces", "releasedAt": "2026-08-10", "source": {"authors": ["Chenxu Du", "Kang An", "Tengyue Wang", "Zhongyu Yang", "Xinqi Yang", "Yuanchi Zhu", "Hebao Zhu", "Ziliang Wang", "Faqiang Qian", "Yunli Yang", "Qibing Ren"], "categories": ["cs.AI"], "id": "2608.09281", "title": "MMArch: Benchmarking Multimodal Reasoning Grounded in Architectural Evidence", "type": "arxiv", "url": "https://arxiv.org/abs/2608.09281"}, "sourceUpdatedAt": "2026-08-10T08:37:14Z", "topics": ["Multimodal", "Reasoning"], "venueAttempts": [], "whyItMatters": "Existing benchmarks test drawing recognition or information extraction, but MMArch evaluates whether models can combine distributed visual evidence with engineering principles, filling a gap in multimodal reasoning evaluation."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
