{"record": {"aliases": ["M$^3$R-Bench: A Unified Benchmark for Evidence-Grounded Multimodal Metaphor Understanding"], "artifacts": [{"id": "arxiv:2608.05817", "kind": "paper", "url": "https://arxiv.org/abs/2608.05817"}, {"id": "gh:hongshi4/m3r-bench", "kind": "repo", "url": "https://github.com/hongshi4/M3R-Bench"}], "authors": ["Hong Jiang", "Junnan Zhu", "Jingwang Huang", "Xiao Sun", "Yuming Yang", "Jiang Zhong", "Ruirui Chen", "Jingman Shi", "Hao Wu", "Nayu Liu", "Xinyi Jiang", "Kaiwen Wei"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2608.05817", "source_url": "https://arxiv.org/abs/2608.05817"}, "categories": ["General AI", "Multimodal"], "description": {"en": "Evaluates unified metaphor understanding across 1,000 image-text instances with annotations for occurrence, Target-Source mapping, sentiment, and stage-wise explanations."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2608.05817", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2608.05817", "title": "M$^3$R-Bench"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "institutions": ["Chongqing University", "Chinese Academy of Sciences", "Institute of Automation", "Agency for Science, Technology and Research", "Institute of Advanced Intelligence and Computing", "Chongqing Medical University", "Tianjin University", "UNSW Sydney"], "institutions_reference": {"basis": "openalex_linked_institutions", "doi": "10.48550/arxiv.2608.05817", "retrieved_at": "2026-11-04", "source_key": "claire-radar:2608.05817", "source_url": "https://openalex.org/W7201849770"}, "key": "claire-radar:2608.05817", "modality": null, "name": "M$^3$R-Bench", "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": "true", "origin_record_id": "bm_m-3-r-bench-a-unified-benchmark-for-eviden_90294861", "origin_source": "arxiv", "recognition_confidence": "0.7", "record_sha256": "e35518b78306f3eb976788f66304a0df74f16766ab118a568a74238df5cfada5", "relation": "introduces", "review_model": "deepseek-v4-pro", "review_state": "ai-reviewed", "reviewed_at": "2026-09-06T08:45:23.913416Z", "source_url": "https://arxiv.org/abs/2608.05817"}, "publisher": null, "released": "2026-08-06", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2608.05817", "source_url": "https://arxiv.org/abs/2608.05817"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2608-05817", "source": "claire_radar", "source_benchmark_id": "2608.05817", "source_metadata": {"claire_radar": {"adoption30d": null, "aliases": ["M$^3$R-Bench: A Unified Benchmark for Evidence-Grounded Multimodal Metaphor Understanding"], "annotation": "Unknown", "applicationDomains": ["General AI"], "area": "Multimodal", "attention": {"asOf": "2026-10-05", "attemptedAt": "2026-10-04T15:31:48.030916Z", "githubRepo": "https://github.com/hongshi4/M3R-Bench", "githubScope": "benchmark_repo", "githubStars": 1, "hfDailySubmittedAt": null, "hfDataset": null, "hfDatasetDownloads": null, "hfDatasetLikes": null, "hfPaperUpvotes": null, "hfPaperUrl": "https://huggingface.co/papers/2608.05817", "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", "state": "unavailable"}}, "status": "fresh"}, "attentionForecast": {"confidence": "Medium", "horizon": "7d", "reason": "The benchmark targets multimodal reasoning with a released dataset and code, and the paper reports results against frontier models, likely attracting NLP and vision-language researchers.", "score": 75}, "capabilities": [], "confidence": "Medium", "construction": "Unknown", "constructionDetail": "Unknown — the indexer does not infer construction details without explicit source evidence.", "copyGeneration": {"generatedAt": "2026-09-06T08:45:23.913416Z", "inputHash": "af5d38e19bb4746db820cf14c8ee2ee722a0f7ae3c1158b87a04bb0c500c5c75", "model": "deepseek-v4-pro"}, "curation": {"canonicalNameEvidence": "M$^3$R-Bench: A Unified Benchmark for Evidence-Grounded Multimodal Metaphor Understanding", "canonicalNameSource": "paper_title", "decisionReason": "The paper and repository provide 1,000 human-verified test instances with code, evaluation scripts, and a reproducible protocol on GitHub.", "model": "deepseek-v4-pro", "reviewedAt": "2026-09-06T08:45:23.913416Z", "state": "ai-reviewed"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "Evaluates unified metaphor understanding across 1,000 image-text instances with annotations for occurrence, Target-Source mapping, sentiment, and stage-wise explanations.", "displayEligible": true, "domainCuration": {"method": "rules-v1", "state": "auto"}, "evaluationMode": "score_submission", "evidence": {"reasonCodes": ["benchmark term in abstract", "benchmark release stated in one sentence", "evaluation protocol evidence", "public artifact URL"], "snippet": "However, existing benchmarks mainly evaluate metaphor understanding through isolated subtasks and lack evidence-grounded explanations, making it difficult to assess whether models establish mappings grounded in visual and textual cues.To address these limitations, we introduce M$^3$R-Bench, a unified and evidence-grounded benchmark containing 1,000 image--text instances with human-verified annotations."}, "familyId": "bmf_2afc09180077", "firstSeenAt": "2026-08-19", "heat": null, "id": "bm_m-3-r-bench-a-unified-benchmark-for-eviden_90294861", "indexedAt": "2026-08-19T11:05:13.395391Z", "industrySectors": [], "links": {"code": "https://github.com/hongshi4/M3R-Bench", "data": null, "paper": "https://arxiv.org/abs/2608.05817", "pdf": "https://arxiv.org/pdf/2608.05817", "project": null, "report": "https://arxiv.org/abs/2608.05817"}, "metrics": [], "motivation": "Metaphor enables the understanding of abstract concepts through cross-domain mappings while conveying affective attitudes.", "name": "M$^3$R-Bench", "oneLine": "Evaluates unified metaphor understanding across 1,000 image-text instances with annotations for occurrence, Target-Source mapping, sentiment, and stage-wise explanations.", "paperTitle": "M$^3$R-Bench: A Unified Benchmark for Evidence-Grounded Multimodal Metaphor Understanding", "primaryDomain": "General AI", "publication": {"evidence": null, "evidenceLevel": "unverified", "evidenceUrl": "https://arxiv.org/abs/2608.05817", "source": "arxiv-metadata", "status": "unverified", "venue": null, "verifiedAt": "2026-08-19T11:05:13.395391Z"}, "publications": [], "publishers": [{"name": "hongshi4", "organizationType": "academic-lab", "role": "benchmark-publisher", "sourceUrl": "https://github.com/hongshi4/M3R-Bench"}], "ranking": {"90d": {"confidence": "Low", "coverage": 0.55, "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": 0.09252268850709358, "value": 1}, "hfDatasetDownloads": {"normalization": "log-max", "normalized": null, "value": null}, "hfPaperUpvotes": {"normalization": "log-max", "normalized": null, "value": null}}, "confidence": "Low", "coverage": 0.55, "rank": 342, "score": 28}, "method": "attention-ranking-v13", "rank": 342, "score": 28}}, "readiness": "Runnable", "recognitionConfidence": 0.7, "relation": "introduces", "releasedAt": "2026-08-06", "source": {"authors": ["Hong Jiang", "Junnan Zhu", "Jingwang Huang", "Xiao Sun", "Yuming Yang", "Jiang Zhong", "Ruirui Chen", "Jingman Shi", "Hao Wu", "Nayu Liu", "Xinyi Jiang", "Kaiwen Wei"], "categories": ["cs.CL"], "id": "2608.05817", "title": "M$^3$R-Bench: A Unified Benchmark for Evidence-Grounded Multimodal Metaphor Understanding", "type": "arxiv", "url": "https://arxiv.org/abs/2608.05817"}, "sourceUpdatedAt": "2026-08-06T09:48:36Z", "topics": ["Multimodal"], "venueAttempts": [], "whyItMatters": "Prior metaphor benchmarks isolate subtasks and lack evidence-grounded explanations, so this benchmark enables assessment of cross-modal reasoning and explanation quality."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
