{"schema_version":"papers/paper-detail-v1","title":"The Scaling Law of Evaluation Failure: Why Simple Averaging Collapses Under Data Sparsity and Item Difficulty Gaps, and How Item Response Theory Recovers Ground Truth Across Domains","surface":"papers","opportunity_kernel":{"paper_id":"9ddfaff9-965b-4292-8ccc-f75190df3fa7","title":"The Scaling Law of Evaluation Failure: Why Simple Averaging Collapses Under Data Sparsity and Item Difficulty Gaps, and How Item Response Theory Recovers Ground Truth Across Domains","authors":["Jung Min Kang"],"arxiv_id":"2605.11205v1","doi":null,"published_at":"2026-05-11T20:17:55.000Z","score_object":{"overall":{"value":4,"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":"2026-05-13T20:19:11.372Z","fresh_until":"2026-06-12T20:19:11.372Z","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":"2026-05-13T20:36:15.254Z","fresh_until":"2026-05-27T20:36:15.254Z","is_stale":true,"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":"2026-05-13T20:19:11.372Z","fresh_until":"2026-06-12T20:19:11.372Z","is_stale":false,"source_count":2,"missingness":[]},"market":{"value":5.9,"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":"2026-05-13T21:03:51.835Z","fresh_until":"2026-05-27T21:03:51.835Z","is_stale":true,"source_count":1,"missingness":[]},"team":{"value":2.7,"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":"2026-05-13T20:19:11.372Z","fresh_until":"2026-05-27T20:19:11.372Z","is_stale":true,"source_count":2,"missingness":[]},"methodology":{"value":3.9,"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":"2026-05-13T20:20:38.351Z","fresh_until":"2026-06-12T20:20:38.351Z","is_stale":false,"source_count":2,"missingness":[]}},"evidence_receipt":{"freshness":"stale","proof_status":"partial","repo_status":"active","references_count":0,"source_count":4,"coverage":0.8333,"missingness":["references"],"unresolved_unknowns":[],"last_verification_at":"2026-05-13T20:36:15.254Z"},"lineage_hash":"9be900d67190f513aff468d3d3bd81fa268c8521d795bd8b153a295839349721"},"distribution":null,"replication_evidence":[],"author_dna":[]}