{"record": {"aliases": [], "artifacts": [{"id": "arxiv:2607.16243", "kind": "paper", "url": "https://arxiv.org/abs/2607.16243"}, {"id": "gh:en-research/robustmad", "kind": "repo", "url": "https://github.com/en-research/RobustMAD"}], "authors": ["Anushiya Arunan", "Xin Li", "Yan Qin", "U-Xuan Tan", "Nhu Khue Vuong", "Xiaoli Li", "Chau Yuen"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2607.16243", "source_url": "https://arxiv.org/abs/2607.16243"}, "categories": ["General AI", "Multimodal", "Robustness", "Safety & Trustworthiness"], "description": {"en": "Evaluates robustness of multimodal small language models for industrial anomaly detection across open-ended queries and visual degradations, using multiple-choice accuracy and LLM-judged open-ended responses."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2607.16243", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2607.16243", "title": "RobustMAD"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "institutions": ["Agency for Science, Technology and Research", "Singapore University of Technology and Design", "Nanyang Technological University", "Chongqing University"], "institutions_reference": {"basis": "openalex_linked_institutions", "doi": "10.48550/arxiv.2607.16243", "retrieved_at": "2026-11-04", "source_key": "claire-radar:2607.16243", "source_url": "https://openalex.org/W7169871234"}, "key": "claire-radar:2607.16243", "modality": null, "name": "RobustMAD", "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_robustmad_0c7403de", "origin_source": "arxiv", "recognition_confidence": "1.0", "record_sha256": "2cb51307f4b8a034704fa7ba62216e526c6976c638385dd3b808c41e1e453b68", "relation": "introduces", "review_state": "unreviewed", "source_url": "https://arxiv.org/abs/2607.16243"}, "publisher": null, "released": "2026-06-26", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2607.16243", "source_url": "https://arxiv.org/abs/2607.16243"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2607-16243", "source": "claire_radar", "source_benchmark_id": "2607.16243", "source_metadata": {"claire_radar": {"adoption30d": null, "aliases": [], "annotation": "Unknown", "applicationDomains": ["General AI"], "area": "Safety & Trustworthiness", "attention": {"asOf": "2026-10-05", "attemptedAt": "2026-10-04T15:31:48.030916Z", "githubRepo": "https://github.com/en-research/RobustMAD", "githubScope": "benchmark_repo", "githubStars": 7, "hfDailySubmittedAt": null, "hfDataset": null, "hfDatasetDownloads": null, "hfDatasetLikes": null, "hfPaperUpvotes": 1, "hfPaperUrl": "https://huggingface.co/papers/2607.16243", "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": "2026-10-04", "state": "stale"}}, "status": "fresh"}, "capabilities": ["Robustness"], "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": "6ea21137d1a73d54364f7629546ff74194bead95c73b6f9b122de37c977e15d3", "model": "deepseek-v4-flash"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "Evaluates robustness of multimodal small language models for industrial anomaly detection across open-ended queries and visual degradations, using multiple-choice accuracy and LLM-judged open-ended responses.", "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": "To address this gap, we develop RobustMAD, the first deployment-motivated benchmark, designed to comprehensively evaluate model robustness through diverse open-ended queries spanning object understanding, anomaly detection, unanswerable problems, and visual quality degradations."}, "familyId": "bmf_191732065a96", "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/en-research/RobustMAD", "url": "https://github.com/en-research/RobustMAD"}, {"basis": "existing-metric-association", "linkedUrl": "https://github.com/en-research/RobustMAD", "url": "https://huggingface.co/papers/2607.16243"}, {"basis": "linked-from-report", "linkedUrl": "https://github.com/en-research/RobustMAD", "url": "https://arxiv.org/abs/2607.16243"}, {"basis": "linked-from-report", "linkedUrl": "https://github.com/en-research/RobustMAD.", "url": "https://arxiv.org/abs/2607.16243"}], "scope": "benchmark_repo", "url": "https://github.com/en-research/RobustMAD", "verification": "source-linked"}], "status": "linked"}, "heat": null, "id": "bm_robustmad_0c7403de", "indexedAt": "2026-08-19T11:05:13.395391Z", "industrySectors": [], "links": {"code": "https://github.com/en-research/RobustMAD", "data": null, "hfPaper": "https://huggingface.co/papers/2607.16243", "paper": "https://arxiv.org/abs/2607.16243", "pdf": "https://arxiv.org/pdf/2607.16243", "project": "https://openreview.net/forum?id=skrA9UYNIZ", "report": "https://arxiv.org/abs/2607.16243"}, "metrics": [], "motivation": "Multimodal industrial anomaly inspection assistants are a critical component of next-generation smart factories, enabling interactive vision-language-based querying.", "name": "RobustMAD", "oneLine": "Evaluates robustness of multimodal small language models for industrial anomaly detection across open-ended queries and visual degradations, using multiple-choice accuracy and LLM-judged open-ended responses.", "paperTitle": "RobustMAD: Evaluating Real-World Robustness of Multimodal Small Language Models for Deployable Anomaly Detection Assistants", "primaryDomain": "General AI", "publication": {"evidence": "Accepted for publication in Transactions on Machine Learning Research (TMLR), 2026 at https://openreview.net/forum?id=skrA9UYNIZ", "evidenceLevel": "author-claim", "evidenceUrl": "https://arxiv.org/abs/2607.16243", "source": "arxiv-comments", "status": "acceptance_claimed", "venue": "publication in Transactions on Machine Learning Research (TMLR), 2026 at https://openreview", "verifiedAt": "2026-10-14T01:46:58.058199Z"}, "publications": [], "publishers": [{"name": "EN Research", "organizationType": "academic-lab", "role": "benchmark-publisher", "sourceUrl": "https://github.com/en-research/RobustMAD"}], "ranking": {}, "readiness": "Runnable", "recognitionConfidence": 1.0, "relation": "introduces", "releasedAt": "2026-06-26", "source": {"authors": ["Anushiya Arunan", "Xin Li", "Yan Qin", "U-Xuan Tan", "Nhu Khue Vuong", "Xiaoli Li", "Chau Yuen"], "categories": ["cs.LG", "cs.AI"], "id": "2607.16243", "title": "RobustMAD: Evaluating Real-World Robustness of Multimodal Small Language Models for Deployable Anomaly Detection Assistants", "type": "arxiv", "url": "https://arxiv.org/abs/2607.16243"}, "sourceUpdatedAt": "2026-06-26T09:15:07Z", "topics": ["Multimodal"], "venueAttempts": [{"decisionRaw": "Accepted for publication in Transactions on Machine Learning Research (TMLR), 2026 at https://openreview.net/forum?id=skrA9UYNIZ", "evidence": [{"level": "author-claim", "observedAt": "2026-10-14T01:46:58.058199Z", "rawValue": "Accepted for publication in Transactions on Machine Learning Research (TMLR), 2026 at https://openreview.net/forum?id=skrA9UYNIZ", "sourceType": "arxiv-comments", "sourceUrl": "https://arxiv.org/abs/2607.16243"}], "reviewStatus": "accepted", "venueName": "publication in Transactions on Machine Learning Research (TMLR), 2026 at https://openreview"}], "whyItMatters": "Assesses deployability of compact models in real-world industrial conditions, identifying failure modes like fragile grounding and hallucination on ill-posed queries."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
