{"record": {"aliases": [], "artifacts": [{"id": "arxiv:2606.26437", "kind": "paper", "url": "https://arxiv.org/abs/2606.26437"}], "authors": ["Siyi Liu", "Aaron Halfaker", "Dan Roth", "Patrick Xia"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2606.26437", "source_url": "https://arxiv.org/abs/2606.26437"}, "categories": ["General AI", "Language & Knowledge", "cs.CL"], "description": {"en": "ConflictBench is a benchmark with ConflictScore metric to quantify how well models acknowledge conflicting evidence in grounding documents, decomposing responses into claims and labeling them against documents."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2606.26437", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2606.26437", "title": "ConflictBench"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "key": "claire-radar:2606.26437", "modality": null, "name": "ConflictBench", "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_conflictbench_d8fc3760", "origin_source": "arxiv", "recognition_confidence": "1.0", "record_sha256": "415e1a3c47f458de9ab69ce8965871616dbde679795209091a000e170eb1bb45", "relation": "introduces", "review_state": "unreviewed", "source_url": "https://arxiv.org/abs/2606.26437"}, "publisher": null, "released": "2026-06-24", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2606.26437", "source_url": "https://arxiv.org/abs/2606.26437"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2606-26437", "source": "claire_radar", "source_benchmark_id": "2606.26437", "source_metadata": {"claire_radar": {"adoption30d": null, "aliases": [], "annotation": "Mixed", "applicationDomains": ["General AI"], "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": 0, "hfPaperUrl": "https://huggingface.co/papers/2606.26437", "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", "lastSuccessfulDate": "2026-10-04", "state": "stale"}}, "status": "stale"}, "capabilities": [], "confidence": "High", "construction": "Aggregate Existing", "constructionDetail": "Unknown — the indexer does not infer construction details without explicit source evidence.", "copyGeneration": {"generatedAt": "2026-08-20T15:22:10.044731Z", "inputHash": "699c153bdab56cf6cd71b2fe4143d3585cbbeaaca8334edfa50d2e2e55c5aa86", "model": "deepseek-v4-flash"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "ConflictBench is a benchmark with ConflictScore metric to quantify how well models acknowledge conflicting evidence in grounding documents, decomposing responses into claims and labeling them against documents.", "domainCuration": {"method": "rules-v1", "state": "auto"}, "evidence": {"reasonCodes": ["exact named benchmark artifact released in abstract", "benchmark release stated in one sentence", "evaluation protocol evidence"], "snippet": "We develop ConflictBench, a benchmark covering diverse forms of conflicts such as ambiguity, contradiction, and divergent opinions, to systematically evaluate our metric."}, "familyId": "bmf_73bdd16dbae8", "firstSeenAt": "2026-08-19", "githubIndex": {"candidateCount": 0, "indexedAt": "2026-09-17T04:40:34.111665+00:00", "repositories": [], "status": "not-found"}, "heat": null, "id": "bm_conflictbench_d8fc3760", "indexedAt": "2026-08-19T11:05:13.395391Z", "industrySectors": [], "links": {"code": null, "data": null, "hfPaper": "https://huggingface.co/papers/2606.26437", "paper": "https://arxiv.org/abs/2606.26437", "pdf": "https://arxiv.org/pdf/2606.26437", "project": null, "report": "https://arxiv.org/abs/2606.26437"}, "metrics": [], "motivation": "Existing metrics for factuality and faithfulness evaluate whether an answer is supported or contradicted by its grounding documents, but they fail to capture when both supporting and contradicting evidence coexist.", "name": "ConflictBench", "oneLine": "ConflictBench is a benchmark with ConflictScore metric to quantify how well models acknowledge conflicting evidence in grounding documents, decomposing responses into claims and labeling them against documents.", "paperTitle": "ConflictScore: Identifying and Measuring How Language Models Handle Conflicting Evidence", "primaryDomain": "General AI", "publication": {"evidence": null, "evidenceLevel": "unverified", "evidenceUrl": "https://arxiv.org/abs/2606.26437", "source": "arxiv-metadata", "status": "unverified", "venue": null, "verifiedAt": "2026-10-14T01:46:58.058199Z"}, "publications": [], "publishers": [{"name": "arXiv", "organizationType": "community", "role": "benchmark-publisher", "sourceUrl": "https://arxiv.org/abs/2606.26437"}], "ranking": {}, "readiness": "Paper only", "recognitionConfidence": 1.0, "relation": "introduces", "releasedAt": "2026-06-24", "source": {"authors": ["Siyi Liu", "Aaron Halfaker", "Dan Roth", "Patrick Xia"], "categories": ["cs.CL", "cs.AI"], "id": "2606.26437", "title": "ConflictScore: Identifying and Measuring How Language Models Handle Conflicting Evidence", "type": "arxiv", "url": "https://arxiv.org/abs/2606.26437"}, "sourceUpdatedAt": "2026-06-24T23:00:09Z", "topics": ["cs.CL"], "venueAttempts": [], "whyItMatters": "Existing factuality metrics ignore coexisting contradictions; ConflictScore provides a nuanced measure over ConflictBench, offering a corrective feedback mechanism for improving truthfulness."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
