{"record": {"aliases": [], "artifacts": [{"id": "arxiv:2609.08174", "kind": "paper", "url": "https://arxiv.org/abs/2609.08174"}], "authors": ["Xiao Yu Cindy Zhang", "Wyeth Wasserman", "Jian Zhu"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2609.08174", "source_url": "https://arxiv.org/abs/2609.08174"}, "categories": ["Biology & Drug Discovery", "Information retrieval", "Language & Knowledge", "cs.AI"], "description": {"en": "A tiered biomedical retrieval benchmark with query-document pairs for concept grounding, relational retrieval, and compositional phenotype-based retrieval."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2609.08174", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2609.08174", "title": "OntologyBench"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "key": "claire-radar:2609.08174", "modality": null, "name": "OntologyBench", "openness": {"code_license": null, "data_license": null, "evidence": [], "status": "unknown"}, "provenance": {"admission_policy_version": "2026-09-09.1", "confidence": "High", "crawl_bundle": "claire_radar_2026-10-14", "crawled_at": "2026-10-14T09:44:51Z", "data_status": "primary-source-indexed", "display_eligible": "false", "origin_record_id": "bm_ontologybench_398d89ca", "origin_source": "arxiv", "recognition_confidence": "1.0", "record_sha256": "770ccb51926baae5b58527d3e6e487fb5b950498e1d6a537bcd3bc5337a7e2c1", "relation": "introduces", "review_model": "deepseek-v4-pro", "review_state": "ai-name-audit-deferred", "reviewed_at": "2026-09-10T00:52:28.558291Z", "source_url": "https://arxiv.org/abs/2609.08174"}, "publisher": null, "released": "2026-09-08", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2609.08174", "source_url": "https://arxiv.org/abs/2609.08174"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2609-08174", "source": "claire_radar", "source_benchmark_id": "2609.08174", "source_metadata": {"claire_radar": {"adoption30d": null, "aliases": [], "annotation": "Unknown", "applicationDomains": ["Biology & Drug Discovery"], "area": "Language & Knowledge", "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/2609.08174", "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"}, "attentionForecast": {"confidence": "Low", "horizon": "7d", "reason": "The biomedical retrieval topic and tiered task structure could attract moderate niche interest, but the lack of visible artifacts or code limits immediate attention.", "score": 32}, "capabilities": ["Information retrieval"], "confidence": "High", "construction": "Unknown", "constructionDetail": "Unknown — the indexer does not infer construction details without explicit source evidence.", "copyGeneration": {"generatedAt": "2026-09-10T00:52:28.558291Z", "inputHash": "87da2f0c066d058ce0aa8be7694d2119e09ed71d0e6d1888f2653d253f14fe7b", "model": "deepseek-v4-pro"}, "curation": {"admissionPolicyVersion": "2026-09-09.1", "canonicalNameEvidence": "OntologyBench: Can Dense Retrieval Satisfy Structured Biomedical Constraints?", "canonicalNameSource": "paper_title", "decisionReason": "The paper introduces the benchmark but no released task inputs, scoring tooling, or public reuse path is evidenced in the supplied links, so its stability and external submission path cannot be confirmed.", "evaluationEvidence": {"reuse": "Fine-tuning on ontology-derived supervision improves performance on several relational and compositional tasks, whereas the evaluated reranking and LLM-based candidate-scoring methods provide little or no end-to-end improvement", "scoring": "embedding performance is generally lower on relational and compositional tasks than on concept-grounding tasks; Fine-tuning on ontology-derived supervision improves performance on several relational and compositional tasks", "task": "a tiered biomedical retrieval benchmark comprising 471,854 training and 125,744 evaluation query-document relevance pairs across concept grounding, relational retrieval, and compositional phenotype-based retrieval"}, "model": "deepseek-v4-pro", "reviewedAt": "2026-09-10T00:52:28.558291Z", "state": "ai-name-audit-deferred"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "A tiered biomedical retrieval benchmark with query-document pairs for concept grounding, relational retrieval, and compositional phenotype-based retrieval.", "displayEligible": false, "domainCuration": {"method": "rules-v1", "state": "auto"}, "evaluationMode": "public_reusable", "evidence": {"reasonCodes": ["coined title prefix ending in Bench or Benchmark", "benchmark release stated in one sentence", "evaluation protocol evidence", "public artifact URL"], "snippet": "We introduce OntologyBench, a tiered biomedical retrieval benchmark comprising 471,854 training and 125,744 evaluation query-document relevance pairs across concept grounding, relational retrieval, and compositional phenotype-based retrieval."}, "familyId": "bmf_0adc82221ef5", "firstSeenAt": "2026-09-09", "heat": null, "id": "bm_ontologybench_398d89ca", "indexedAt": "2026-09-09T06:18:29.794093Z", "industrySectors": ["Pharma & Biotech"], "links": {"code": null, "data": null, "paper": "https://arxiv.org/abs/2609.08174", "pdf": "https://arxiv.org/pdf/2609.08174", "project": "https://openreview.net/forum?id=RMiOTIub1j&noteId=RMiOTIub1j", "report": "https://arxiv.org/abs/2609.08174"}, "metrics": [], "motivation": "We introduce OntologyBench, a tiered biomedical retrieval benchmark comprising 471,854 training and 125,744 evaluation query-document relevance pairs across concept grounding, relational retrieval, and compositional phenotype-based retrieval.", "name": "OntologyBench", "oneLine": "A tiered biomedical retrieval benchmark with query-document pairs for concept grounding, relational retrieval, and compositional phenotype-based retrieval.", "paperTitle": "OntologyBench: Can Dense Retrieval Satisfy Structured Biomedical Constraints?", "primaryDomain": "Biology & Drug Discovery", "publication": {"evidence": "19 pages and 1 figure. Accepted to EMNLP 2026 Main conference. https://openreview.net/forum?id=RMiOTIub1j&noteId=RMiOTIub1j", "evidenceLevel": "author-claim", "evidenceUrl": "https://arxiv.org/abs/2609.08174", "source": "arxiv-comments", "status": "acceptance_claimed", "venue": "EMNLP 2026 Main conference", "verifiedAt": "2026-09-09T06:18:29.794093Z"}, "publications": [], "ranking": {"30d": {"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": 30, "historyDate": "2026-09-05", "rank": null, "score": null, "windowDays": 30}, "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}, "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": "2026-09-08", "source": {"authors": ["Xiao Yu Cindy Zhang", "Wyeth Wasserman", "Jian Zhu"], "categories": ["cs.AI"], "id": "2609.08174", "title": "OntologyBench: Can Dense Retrieval Satisfy Structured Biomedical Constraints?", "type": "arxiv", "url": "https://arxiv.org/abs/2609.08174"}, "sourceUpdatedAt": "2026-09-08T03:04:36Z", "topics": ["cs.AI"], "venueAttempts": [{"decisionRaw": "19 pages and 1 figure. Accepted to EMNLP 2026 Main conference. https://openreview.net/forum?id=RMiOTIub1j&noteId=RMiOTIub1j", "evidence": [{"level": "author-claim", "observedAt": "2026-09-09T06:18:29.794093Z", "rawValue": "19 pages and 1 figure. Accepted to EMNLP 2026 Main conference. https://openreview.net/forum?id=RMiOTIub1j&noteId=RMiOTIub1j", "sourceType": "arxiv-comments", "sourceUrl": "https://arxiv.org/abs/2609.08174"}], "reviewStatus": "accepted", "venueName": "EMNLP 2026 Main conference"}], "whyItMatters": "It targets the gap in evaluating dense retrieval under structured biomedical constraints, where models struggle with relational and compositional tasks."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
