{"record": {"aliases": [], "artifacts": [{"id": "arxiv:2608.09772", "kind": "paper", "url": "https://arxiv.org/abs/2608.09772"}], "authors": ["Zhanna Mukhametsharip", "Vera Demberg", "Varsha Suresh"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2608.09772", "source_url": "https://arxiv.org/abs/2608.09772"}, "categories": ["General AI", "Multimodal"], "description": {"en": "PragMatch is a controlled set of 3,000 image-text pairs derived from MMSD2.0 for studying pragmatic incongruity in multimodal sarcasm detection, with original sarcastic examples and constructed literal and hard-negative pairs."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2608.09772", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2608.09772", "title": "PragMatch"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "institutions": ["Saarland University", "Max Planck Institute for Informatics"], "institutions_reference": {"basis": "openalex_linked_institutions", "doi": "10.48550/arxiv.2608.09772", "retrieved_at": "2026-11-04", "source_key": "claire-radar:2608.09772", "source_url": "https://openalex.org/W7202167582"}, "key": "claire-radar:2608.09772", "modality": null, "name": "PragMatch", "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_pragmatch_5cfd2fc5", "origin_source": "arxiv", "recognition_confidence": "1.0", "record_sha256": "30bf1171192517a97f5ac85e005d4ec20d96e30d6c5f8cd5a767ffa72c19b367", "relation": "introduces", "review_state": "unreviewed", "source_url": "https://arxiv.org/abs/2608.09772"}, "publisher": null, "released": "2026-08-10", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2608.09772", "source_url": "https://arxiv.org/abs/2608.09772"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2608-09772", "source": "claire_radar", "source_benchmark_id": "2608.09772", "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/2608.09772", "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": [], "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": "02b3858c0d5e91311baa7320036df25cb670bac676f468442886c33674339fa7", "model": "deepseek-v4-flash"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "PragMatch is a controlled set of 3,000 image-text pairs derived from MMSD2.0 for studying pragmatic incongruity in multimodal sarcasm detection, with original sarcastic examples and constructed literal and hard-negative pairs.", "domainCuration": {"method": "rules-v1", "state": "auto"}, "evidence": {"reasonCodes": ["exact coined title identity tied to benchmark evidence", "benchmark release stated in one sentence"], "snippet": "We introduce PragMatch, a controlled benchmark of 3,000 image-text pairs derived from MMSD2.0, including original sarcastic examples and constructed literal and hard-negative pairs."}, "familyId": "bmf_c5b11c599fa8", "firstSeenAt": "2026-08-19", "githubIndex": {"candidateCount": 0, "indexedAt": "2026-09-17T04:40:34.111665+00:00", "repositories": [], "status": "not-found"}, "heat": null, "id": "bm_pragmatch_5cfd2fc5", "indexedAt": "2026-08-19T11:05:13.395391Z", "industrySectors": [], "links": {"code": null, "data": null, "hfPaper": "https://huggingface.co/papers/2608.09772", "paper": "https://arxiv.org/abs/2608.09772", "pdf": "https://arxiv.org/pdf/2608.09772", "project": null, "report": "https://arxiv.org/abs/2608.09772"}, "metrics": [], "motivation": "Large Vision-Language Models (LVLMs) have demonstrated strong performance on multimodal benchmarks, yet it remains unclear whether they genuinely reason about relationships between images and text or rely on superficial correlations, known as shortcut learning.", "name": "PragMatch", "oneLine": "PragMatch is a controlled set of 3,000 image-text pairs derived from MMSD2.0 for studying pragmatic incongruity in multimodal sarcasm detection, with original sarcastic examples and constructed literal and hard-negative pairs.", "paperTitle": "PragMatch: Separating Pragmatic Incongruity from Cross-Modal Mismatch in Large Vision-Language Models", "primaryDomain": "General AI", "publication": {"evidence": null, "evidenceLevel": "unverified", "evidenceUrl": "https://arxiv.org/abs/2608.09772", "source": "arxiv-metadata", "status": "unverified", "venue": null, "verifiedAt": "2026-10-14T01:46:58.058199Z"}, "publications": [], "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": "Paper only", "recognitionConfidence": 1.0, "relation": "introduces", "releasedAt": "2026-08-10", "source": {"authors": ["Zhanna Mukhametsharip (1)", "Vera Demberg (1 and 2)", "Varsha Suresh (2) ((1) Saarland University", "Germany", "(2) Max Planck Institute for Informatics", "Germany)"], "categories": ["cs.CL"], "id": "2608.09772", "title": "PragMatch: Separating Pragmatic Incongruity from Cross-Modal Mismatch in Large Vision-Language Models", "type": "arxiv", "url": "https://arxiv.org/abs/2608.09772"}, "sourceUpdatedAt": "2026-08-10T16:00:04Z", "topics": ["Multimodal"], "venueAttempts": [], "whyItMatters": "This resource helps investigate whether large vision-language models rely on superficial cues instead of genuine reasoning in multimodal sarcasm, highlighting practical limitations in model evaluation."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
