{"record": {"aliases": [], "artifacts": [{"id": "arxiv:2608.26382", "kind": "paper", "url": "https://arxiv.org/abs/2608.26382"}, {"id": "gh:mahmoodlab/viper", "kind": "repo", "url": "https://github.com/mahmoodlab/viper"}], "authors": ["Luca L. Weishaupt", "Simone de Brot", "Javier Asin", "Llorenç Grau-Roma", "Nic G. Reitsam", "Andrew H. Song", "Dongmin Bang", "Stefan T. Kaluziak", "Long Phi Le", "Jakob Nikolas Kather", "Faisal Mahmood", "Guillaume Jaume"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2608.26382", "source_url": "https://arxiv.org/abs/2608.26382"}, "categories": ["Biology & Drug Discovery", "Multimodal"], "description": {"en": "VIPER evaluates vision-language models on toxicologic pathology via 1,251 questions over 419 H&E-stained rat histology images across organ systems, covering multiple-choice, KPrim, and free-text formats with domain-expert validation."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2608.26382", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2608.26382", "title": "VIPER"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "institutions": ["University of Bern", "University of California, Davis", "University of Augsburg", "Mass General Brigham", "The University of Texas MD Anderson Cancer Center", "Harvard University", "Technische Universität Dresden", "University of Lausanne"], "institutions_reference": {"basis": "openalex_linked_institutions", "doi": "10.48550/arxiv.2608.26382", "retrieved_at": "2026-11-04", "source_key": "claire-radar:2608.26382", "source_url": "https://openalex.org/W7204591369"}, "key": "claire-radar:2608.26382", "modality": null, "name": "VIPER", "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": "true", "origin_record_id": "bm_viper_dbe65c04", "origin_source": "arxiv", "recognition_confidence": "1.0", "record_sha256": "88a5d8d5dd49052d85572e6c78268980852185cb82877d453b817affb3d32f77", "relation": "introduces", "review_model": "deepseek-v4-pro", "review_state": "ai-reviewed", "reviewed_at": "2026-09-01T09:31:26.845990Z", "source_url": "https://arxiv.org/abs/2608.26382"}, "publisher": null, "released": "2026-08-26", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2608.26382", "source_url": "https://arxiv.org/abs/2608.26382"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2608-26382", "source": "claire_radar", "source_benchmark_id": "2608.26382", "source_metadata": {"claire_radar": {"adoption30d": null, "aliases": [], "annotation": "Unknown", "applicationDomains": ["Biology & Drug Discovery"], "area": "Multimodal", "attention": {"asOf": "2026-10-05", "attemptedAt": "2026-10-04T15:31:48.030916Z", "githubRepo": "https://github.com/mahmoodlab/viper", "githubScope": "benchmark_repo", "githubStars": 10, "hfDailySubmittedAt": null, "hfDataset": null, "hfDatasetDownloads": null, "hfDatasetLikes": null, "hfPaperUpvotes": 1, "hfPaperUrl": "https://huggingface.co/papers/2608.26382", "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": "2026-09-24", "state": "stale"}}, "status": "fresh"}, "attentionForecast": {"confidence": "Medium", "horizon": "7d", "reason": "The benchmark addresses a niche but novel domain and includes public code and leaderboard-style results, likely to draw moderate attention from the pathology and vision-language modeling communities.", "score": 45}, "capabilities": [], "confidence": "High", "construction": "Transform Existing", "constructionDetail": "Unknown — the indexer does not infer construction details without explicit source evidence.", "copyGeneration": {"generatedAt": "2026-09-01T09:31:26.845990Z", "inputHash": "90336eced82d0d6af64c48106b01085ccb38d6e9c203e107171afcdc44e87604", "model": "deepseek-v4-pro"}, "curation": {"canonicalNameEvidence": "VIPER: An Expert-Curated Benchmark for Vision-Language Models in Veterinary Pathology", "canonicalNameSource": "paper_title", "decisionReason": "The paper explicitly introduces VIPER as an expert-curated benchmark, provides a detailed evaluation protocol with reproducible scoring, and public code and data availability, fulfilling criteria for a published benchmark.", "model": "deepseek-v4-pro", "reviewedAt": "2026-09-01T09:31:26.845990Z", "state": "ai-reviewed"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "VIPER evaluates vision-language models on toxicologic pathology via 1,251 questions over 419 H&E-stained rat histology images across organ systems, covering multiple-choice, KPrim, and free-text formats with domain-expert validation.", "displayEligible": true, "domainCuration": {"method": "rules-v1", "state": "auto"}, "evaluationMode": "score_submission", "evidence": {"reasonCodes": ["exact coined title identity tied to benchmark evidence", "benchmark release stated in one sentence", "public artifact URL"], "snippet": "To address it, we introduce VIPER, the first expert-curated benchmark for vision-language model evaluation in toxicologic pathology."}, "familyId": "bmf_895d1ba9ca06", "firstSeenAt": "2026-08-29", "heat": null, "id": "bm_viper_dbe65c04", "indexedAt": "2026-08-29T08:02:23.506802Z", "industrySectors": ["Pharma & Biotech"], "links": {"code": "https://github.com/mahmoodlab/viper", "data": null, "paper": "https://arxiv.org/abs/2608.26382", "pdf": "https://arxiv.org/pdf/2608.26382", "project": null, "report": "https://arxiv.org/abs/2608.26382"}, "metrics": [], "motivation": "Pathology vision-language models are advancing rapidly, yet existing benchmarks remain focused on human tissue, particularly oncology, leaving non-human pathology largely unaddressed.", "name": "VIPER", "oneLine": "VIPER evaluates vision-language models on toxicologic pathology via 1,251 questions over 419 H&E-stained rat histology images across organ systems, covering multiple-choice, KPrim, and free-text formats with domain-expert validation.", "paperTitle": "VIPER: An Expert-Curated Benchmark for Vision-Language Models in Veterinary Pathology", "primaryDomain": "Biology & Drug Discovery", "publication": {"evidence": null, "evidenceLevel": "unverified", "evidenceUrl": "https://arxiv.org/abs/2608.26382", "source": "arxiv-metadata", "status": "unverified", "venue": null, "verifiedAt": "2026-08-29T08:02:23.506802Z"}, "publications": [], "publishers": [{"name": "Mahmood Lab", "organizationType": "academic-lab", "role": "benchmark-publisher", "sourceUrl": "https://github.com/mahmoodlab/viper"}], "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": 0.32007591406276925, "value": 10}, "hfDatasetDownloads": {"normalization": "log-max", "normalized": null, "value": null}, "hfPaperUpvotes": {"normalization": "log-max", "normalized": 0.12247537038455211, "value": 1}}, "confidence": "Medium", "coverage": 0.7, "rank": 192, "score": 34}, "method": "attention-ranking-v13", "rank": 192, "score": 34}}, "readiness": "Runnable", "recognitionConfidence": 1.0, "relation": "introduces", "releasedAt": "2026-08-26", "source": {"authors": ["Luca L. Weishaupt", "Simone de Brot", "Javier Asin", "Lloren\\c{c} Grau-Roma", "Nic G. Reitsam", "Andrew H. Song", "Dongmin Bang", "Stefan T. Kaluziak", "Long Phi Le", "Jakob Nikolas Kather", "Faisal Mahmood", "Guillaume Jaume"], "categories": ["cs.CV"], "id": "2608.26382", "title": "VIPER: An Expert-Curated Benchmark for Vision-Language Models in Veterinary Pathology", "type": "arxiv", "url": "https://arxiv.org/abs/2608.26382"}, "sourceUpdatedAt": "2026-08-26T20:12:27Z", "topics": ["Multimodal"], "venueAttempts": [], "whyItMatters": "VIPER addresses the lack of evaluation for non-human pathology by providing an expert-curated benchmark for toxicologic pathology, offering comparative scores across 16 models and revealing domain gaps and over-diagnosis risks."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
