{"schema_version":"papers/paper-detail-v1","title":"Teaching Language Models to Check Grounded Claim Factuality with Human Test-Taking Strategies","surface":"papers","opportunity_kernel":{"paper_id":"c60b4133-449d-4838-b106-084472b2a51a","title":"Teaching Language Models to Check Grounded Claim Factuality with Human Test-Taking Strategies","authors":["Yuxuan Ye","Raul Santos-Rodriguez","Edwin Simpson"],"arxiv_id":"2605.29712v1","doi":null,"published_at":"2026-05-28T10:11:42.000Z","score_object":{"overall":{"value":7,"scale":"0-10","confidence":0.85,"confidence_reason":"Backfilled from persisted papers.viability_score.","model_version":"phase0-backfill-v1","pipeline_version":"phase0-kernel-v1","computed_at":"2026-05-29T20:13:12.447Z","fresh_until":"2026-06-28T20:13:12.447Z","is_stale":false,"source_count":1,"missingness":[]},"technical":{"value":1.4,"scale":"0-10","confidence":0.55,"confidence_reason":"Backfilled from paper_extraction_scorecards.reconstruction_score.","model_version":"phase0-backfill-v1","pipeline_version":"phase0-kernel-v1","computed_at":"2026-05-29T20:34:21.217Z","fresh_until":"2026-06-12T20:34:21.217Z","is_stale":false,"source_count":1,"missingness":["reproducibility_results.reproducibility_score","deployability_scores.score"]},"commercial":{"value":5,"scale":"0-10","confidence":0.75,"confidence_reason":"Backfilled from persisted commercial_flags and repo availability.","model_version":"phase0-backfill-v1","pipeline_version":"phase0-kernel-v1","computed_at":"2026-05-29T20:13:12.447Z","fresh_until":"2026-06-28T20:13:12.447Z","is_stale":false,"source_count":2,"missingness":[]},"market":{"value":5.5,"scale":"0-10","confidence":0.45,"confidence_reason":"Heuristic market score materialized from paper metadata, repo availability, and deployability.","model_version":"phase0-backfill-v1","pipeline_version":"phase0-kernel-v1","computed_at":"2026-05-29T20:45:10.463Z","fresh_until":"2026-06-12T20:45:10.463Z","is_stale":false,"source_count":1,"missingness":[]},"team":{"value":4.55,"scale":"0-10","confidence":0.42,"confidence_reason":"Heuristic fallback from paper author count and extracted affiliations.","model_version":"phase0-backfill-v1","pipeline_version":"phase0-kernel-v1","computed_at":"2026-05-29T20:13:12.447Z","fresh_until":"2026-06-12T20:13:12.447Z","is_stale":false,"source_count":2,"missingness":[]},"methodology":{"value":4.7,"scale":"0-10","confidence":0.82,"confidence_reason":"Backfilled from paper_extraction_scorecards.total_score.","model_version":"phase0-backfill-v1","pipeline_version":"phase0-kernel-v1","computed_at":"2026-05-29T20:19:12.836Z","fresh_until":"2026-06-28T20:19:12.836Z","is_stale":false,"source_count":2,"missingness":[]}},"evidence_receipt":{"freshness":"fresh","proof_status":"unverified","repo_status":"active","references_count":0,"source_count":4,"coverage":0.8333,"missingness":["references"],"unresolved_unknowns":[],"last_verification_at":"2026-05-29T20:34:21.217Z"},"lineage_hash":"1ce93930972f89024f93f9b4892a4f37cbb6cac2990d823031ddcd5b54d04870"},"distribution":null,"replication_evidence":[],"author_dna":[]}