{"record": {"aliases": [], "artifacts": [{"id": "arxiv:2607.27637", "kind": "paper", "url": "https://arxiv.org/abs/2607.27637"}, {"id": "gh:zhuwenjie98/mmooc", "kind": "repo", "url": "https://github.com/ZhuWenjie98/MMOOC"}, {"kind": "website", "url": "https://zhuwenjie98.github.io/MMOOC-project-page/"}], "authors": ["Wenjie Zhu", "Yabin Zhang", "Wenjun Zeng", "Lei Zhang"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2607.27637", "source_url": "https://arxiv.org/abs/2607.27637"}, "categories": ["General AI", "Multimodal"], "description": {"en": "MMOOC evaluates multimodal large language models on out-of-context (OOC) and shifted in-context (Shifted IC) visual question answering. It contains over 41K image-question pairs covering three question formats, eight shift types, and six visual scenarios. Responses are scored for accuracy and refusal rate, with an LLM-as-a-judge metric for reasoning correctness."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2607.27637", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2607.27637", "title": "MMOOC"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "institutions": ["Hong Kong Polytechnic University", "Harbin Institute of Technology", "Eastern Institute of Technology, Ningbo"], "institutions_reference": {"basis": "openalex_linked_institutions", "doi": "10.48550/arxiv.2607.27637", "retrieved_at": "2026-11-04", "source_key": "claire-radar:2607.27637", "source_url": "https://openalex.org/W7171924854"}, "key": "claire-radar:2607.27637", "modality": null, "name": "MMOOC", "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_mmooc_7beeb312", "origin_source": "arxiv", "recognition_confidence": "1.0", "record_sha256": "d0ac1f622e8ff7a320341b0120ecf727fc9ab4e2e3af6b601780d62a6b1dd0a6", "relation": "introduces", "review_state": "unreviewed", "source_url": "https://arxiv.org/abs/2607.27637"}, "publisher": null, "released": "2026-07-30", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2607.27637", "source_url": "https://arxiv.org/abs/2607.27637"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2607-27637", "source": "claire_radar", "source_benchmark_id": "2607.27637", "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": "https://github.com/ZhuWenjie98/MMOOC", "githubScope": "benchmark_repo", "githubStars": 31, "hfDailySubmittedAt": "2026-08-11T00:00:00.000Z", "hfDataset": null, "hfDatasetDownloads": null, "hfDatasetLikes": null, "hfPaperUpvotes": 6, "hfPaperUrl": "https://huggingface.co/papers/2607.27637", "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": {"observedAt": "2026-10-04T15:31:48.030916Z", "state": "fresh"}}, "status": "fresh"}, "capabilities": [], "confidence": "High", "construction": "Transform Existing", "constructionDetail": "Unknown — the indexer does not infer construction details without explicit source evidence.", "copyGeneration": {"generatedAt": "2026-08-20T15:22:10.044731Z", "inputHash": "185953f78351ad4e6fd1aec2a3adc8e362b4ffa5bcc00ab38aa66c6f6989464e", "model": "deepseek-v4-flash"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "MMOOC evaluates multimodal large language models on out-of-context (OOC) and shifted in-context (Shifted IC) visual question answering. It contains over 41K image-question pairs covering three question formats, eight shift types, and six visual scenarios. Responses are scored for accuracy and refusal rate, with an LLM-as-a-judge metric for reasoning correctness.", "domainCuration": {"method": "rules-v1", "state": "auto"}, "evidence": {"reasonCodes": ["exact coined title identity tied to benchmark evidence", "benchmark release stated in one sentence", "evaluation protocol evidence", "public artifact URL"], "snippet": "To fill this gap, we present MMOOC, a large-scale benchmark for evaluating refusal and robust answering abilities of MLLMs."}, "familyId": "bmf_4c8571bb6649", "firstSeenAt": "2026-08-19", "githubIndex": {"candidateCount": 1, "indexedAt": "2026-09-17T04:40:34.111665+00:00", "repositories": [{"evidence": [{"basis": "existing-code-link", "linkedUrl": "https://github.com/ZhuWenjie98/MMOOC", "url": "https://github.com/ZhuWenjie98/MMOOC"}, {"basis": "existing-metric-association", "linkedUrl": "https://github.com/ZhuWenjie98/MMOOC", "url": "https://huggingface.co/papers/2607.27637"}, {"basis": "linked-from-project", "linkedUrl": "https://github.com/ZhuWenjie98/MMOOC", "url": "https://zhuwenjie98.github.io/MMOOC-project-page/"}], "scope": "benchmark_repo", "url": "https://github.com/ZhuWenjie98/MMOOC", "verification": "source-linked"}], "status": "linked"}, "heat": null, "id": "bm_mmooc_7beeb312", "indexedAt": "2026-08-19T11:05:13.395391Z", "industrySectors": [], "links": {"code": "https://github.com/ZhuWenjie98/MMOOC", "data": null, "hfPaper": "https://huggingface.co/papers/2607.27637", "paper": "https://arxiv.org/abs/2607.27637", "pdf": "https://arxiv.org/pdf/2607.27637", "project": "https://zhuwenjie98.github.io/MMOOC-project-page/", "report": "https://arxiv.org/abs/2607.27637"}, "metrics": [], "motivation": "Multimodal Large Language Models (MLLMs) have achieved strong performance on a wide range of vision-language tasks, but often fail under imperfect or shifted contexts.", "name": "MMOOC", "oneLine": "MMOOC evaluates multimodal large language models on out-of-context (OOC) and shifted in-context (Shifted IC) visual question answering. It contains over 41K image-question pairs covering three question formats, eight shift types, and six visual scenarios. Responses are scored for accuracy and refusal rate, with an LLM-as-a-judge metric for reasoning correctness.", "paperTitle": "MMOOC: A Comprehensive Benchmark for Out-of-Context Evaluation in Multimodal Large Language Models", "primaryDomain": "General AI", "publication": {"evidence": null, "evidenceLevel": "unverified", "evidenceUrl": "https://arxiv.org/abs/2607.27637", "source": "arxiv-metadata", "status": "unverified", "venue": null, "verifiedAt": "2026-10-14T01:46:58.058199Z"}, "publications": [], "publishers": [{"name": "MMOOC Project Team", "organizationType": "academic-lab", "role": "benchmark-publisher", "sourceUrl": "https://zhuwenjie98.github.io/MMOOC-project-page/"}], "ranking": {"90d": {"confidence": "Medium", "coverage": 0.7, "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": 0.46261344253546793, "value": 31}, "hfDatasetDownloads": {"normalization": "log-max", "normalized": null, "value": null}, "hfPaperUpvotes": {"normalization": "log-max", "normalized": 0.34383183387990046, "value": 6}}, "confidence": "Medium", "coverage": 0.7, "rank": 100, "score": 46}, "method": "attention-ranking-v13", "rank": 100, "score": 46}}, "readiness": "Runnable", "recognitionConfidence": 1.0, "relation": "introduces", "releasedAt": "2026-07-30", "source": {"authors": ["Wenjie Zhu", "Yabin Zhang", "Wenjun Zeng", "Lei Zhang"], "categories": ["cs.CV"], "id": "2607.27637", "title": "MMOOC: A Comprehensive Benchmark for Out-of-Context Evaluation in Multimodal Large Language Models", "type": "arxiv", "url": "https://arxiv.org/abs/2607.27637"}, "sourceUpdatedAt": "2026-08-01T04:59:28Z", "topics": ["Multimodal"], "venueAttempts": [], "whyItMatters": "Existing benchmarks focus on unanswerable questions but overlook answerable shifted contexts. MMOOC provides a joint evaluation of refusal and robust answering, offering insight into model reliability in real-world scenarios where contexts are imperfect. This supports comparing models on a balanced measure of capability and safety."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
