{"id":"W4411306282","doi":"10.1007/s41666-025-00204-w","title":"LongHealth: A Question Answering Benchmark with Long Clinical Documents","year":2025,"lang":"en","type":"article","venue":"Journal of Healthcare Informatics Research","topic":"Topic Modeling","field":"Computer Science","cited_by":4,"is_retracted":false,"has_abstract":true,"ca_institutions":"Artificial Intelligence in Medicine (Canada)","funders":"Technische Universität München","keywords":"Benchmark (surveying); Computer science; Sorting; Identification (biology); Unstructured data; Data science; Information retrieval; Data mining; Artificial intelligence; Natural language processing; Big data; Programming language","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.01128923,0.0001146806,0.0003422039,0.0008197504,0.000285134,0.0003135382,0.001195324,0.0001297572,0.000004270281],"category_scores_gemma":[0.0005215118,0.00008670057,0.00007765541,0.0009766804,0.00008319865,0.001442439,0.0003460427,0.001661701,0.000009952631],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0004261761,"about_ca_system_score_gemma":0.002341365,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001752261,"about_ca_topic_score_gemma":0.0001068339,"domain_scores_codex":[0.9954358,0.0005345391,0.001820333,0.0001317043,0.001502341,0.0005752822],"domain_scores_gemma":[0.996187,0.0006071463,0.0005079438,0.000562689,0.001799786,0.0003354143],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0002743762,0.0002143372,0.1023696,0.002285489,0.0001409571,0.0001994487,0.005246343,0.002285998,0.000003663741,0.08190624,0.002584284,0.8024893],"study_design_scores_gemma":[0.008900749,0.009548436,0.1244103,0.01621321,0.00004890899,0.001711997,0.007343581,0.7345421,0.0003306077,0.04160013,0.05422758,0.001122324],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1685676,0.0007725991,0.8099602,0.01687062,0.0008069357,0.0004089846,7.862544e-7,0.00003317578,0.00257914],"genre_scores_gemma":[0.8709603,0.0007278232,0.1272406,0.0007630303,0.0001438848,0.00000580054,8.858184e-7,0.000006660886,0.0001510035],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.801367,"threshold_uncertainty_score":0.7219355,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1073749660151189,"score_gpt":0.5006962605306308,"score_spread":0.3933212945155119,"validation_status":"score_only:v0-immature-baseline","note":"Baseline scores from an immature model (maturity gate not passed). Scores rank; they never assert a category."}}