{"bootstrap_samples":2000,"confidence_interval_95":[0.7961746266338624,0.8156941944618167],"evaluator_portability_patch":"candidate relative paths normalized to POSIX separators on Windows","evaluator_revision":"cfa88c1eb1c2ec4495c84d6820ffe85d33b7408c","evaluator_tracked_patch_sha256":"sha256:e93e8b7ba7f730d416def8629cd54dfd1f3925c8814bc72aaeb5fa4f20b2e734","evaluator_worktree_clean":false,"failure_evidence_sha256":"sha256:6d92c7ba23c7716357c8d6e4cb928171e2a2df3fc76b3078f752e1301589f8b5","html_report_sha256":"sha256:4487890a8f25febf14570a2c46db32a960a31d4df5e77f7806ab7ff089183154","input_count":1403,"overall_score":0.8060068382873822,"per_jsonl":{"arxiv_math.jsonl":{"pass_rate":0.8558250768705159,"passed":2505,"total":2927},"baseline":{"pass_rate":0.9899569583931134,"passed":1380,"total":1394},"headers_footers.jsonl":{"pass_rate":0.95,"passed":722,"total":760},"long_tiny_text.jsonl":{"pass_rate":0.7941176470588235,"passed":351,"total":442},"multi_column.jsonl":{"pass_rate":0.8122171945701357,"passed":718,"total":884},"old_scans.jsonl":{"pass_rate":0.3688212927756654,"passed":194,"total":526},"old_scans_math.jsonl":{"pass_rate":0.7925764192139738,"passed":363,"total":458},"table_tests.jsonl":{"pass_rate":0.8845401174168297,"passed":904,"total":1022}},"rule_failure_count":1276,"schema":"folynta.olmocr-official-evaluation.v1","source_manifest_sha256":"sha256:6af470c73fcf9e164c418ca49b81cc92d0e60199f9dcb714cb634339c22a3865","test_count":8413,"type_breakdown":{"absent":{"pass_rate":0.945321992709599,"test_count":823},"baseline":{"pass_rate":0.9900213827512473,"test_count":1403},"math":{"pass_rate":0.8472673559822748,"test_count":3385},"order":{"pass_rate":0.7219604147031102,"test_count":1061},"present":{"pass_rate":0.6019417475728155,"test_count":721},"table":{"pass_rate":0.884313725490196,"test_count":1020}}}
