{"id":"W7106854864","doi":"10.48448/fd2y-y664","title":"ScholarBench: A Bilingual Benchmark for Abstraction, Comprehension, and Reasoning Evaluation in Academic Contexts","year":2025,"lang":"","type":"other","venue":"Open MIND","topic":"","field":"","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Kootenay Association for Science & Technology","funders":"","keywords":"Benchmark (surveying); Scalability; Normative; Model-based reasoning; Computational linguistics","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.007325583,0.003433226,0.001090741,0.006814618,0.00159005,0.004506003,0.004214175,0.00292679,0.009818954],"category_scores_gemma":[0.03743741,0.0005642543,0.001788099,0.005257592,0.00105519,0.005910873,0.005050438,0.002108997,0.008087963],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002720308,"about_ca_system_score_gemma":0.004407271,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01669662,"about_ca_topic_score_gemma":0.03216414,"domain_scores_codex":[0.9906676,0.003920717,0.001045203,0.001931812,0.001793977,0.0006407253],"domain_scores_gemma":[0.9843929,0.007235453,0.0007924636,0.002687938,0.003633982,0.001257199],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.00207339,0.003449866,0.02925976,0.005984468,0.0007853291,0.0008264653,0.001805796,0.04157621,0.007859559,0.01596623,0.5082505,0.3821624],"study_design_scores_gemma":[0.001963423,0.002021249,0.03715274,0.0019832,0.0005255296,0.001157172,0.004781232,0.4608009,0.03370779,0.04367023,0.4117923,0.0004442711],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"methods","genre_scores_codex":[0.4090337,0.01575151,0.1454265,0.004499975,0.00191559,0.003505204,0.2202294,0.08013666,0.1195014],"genre_scores_gemma":[0.2647861,0.001497299,0.1738181,0.001059907,0.0001898485,0.002242159,0.5440937,0.002839874,0.00947307],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.01669662,"threshold_uncertainty_score":0.03874189,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.09136926799322435,"score_gpt":0.4277614164705215,"score_spread":0.3363921484772971,"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."}}