{"id":"W4409347925","doi":"10.1609/aaai.v39i23.34673","title":"Evaluating LLM Reasoning in the Operations Research Domain with ORQA","year":2025,"lang":"en","type":"article","venue":"Proceedings of the AAAI Conference on Artificial Intelligence","topic":"Business Process Modeling and Analysis","field":"Business, Management and Accounting","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of British Columbia; Huawei Technologies (Canada)","funders":"","keywords":"Domain (mathematical analysis); Computer science; Management science; Mathematics; Engineering","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":[0.02609639,0.001704472,0.0009842414,0.004574101,0.0008577637,0.004427287,0.00304637,0.002838955,0.004568594],"category_scores_gemma":[0.1306716,0.0004620782,0.001865741,0.003408439,0.001647139,0.006999018,0.00451665,0.002920685,0.001275454],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.003204545,"about_ca_system_score_gemma":0.003864278,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01229933,"about_ca_topic_score_gemma":0.01321948,"domain_scores_codex":[0.9791829,0.01261475,0.001590378,0.002686945,0.003394735,0.0005301687],"domain_scores_gemma":[0.8832118,0.09178621,0.005303788,0.01212976,0.00573132,0.001837085],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00177867,0.002190696,0.05834325,0.003298842,0.001082619,0.0005103541,0.002163117,0.5281911,0.004914233,0.0568512,0.04447776,0.2961982],"study_design_scores_gemma":[0.0002385198,0.0002968195,0.005216276,0.0002606897,0.0001043498,0.0001487824,0.000718595,0.9123701,0.003116585,0.0635201,0.01395478,0.00005440595],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.4645884,0.007277479,0.4328543,0.01481285,0.0005890517,0.001570913,0.02404205,0.02211458,0.03215042],"genre_scores_gemma":[0.677668,0.001145039,0.2848484,0.002443499,0.000251589,0.0004802633,0.03118823,0.0004637352,0.001511217],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.02609639,"threshold_uncertainty_score":0.1380125,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2101254107643576,"score_gpt":0.40640091708382,"score_spread":0.1962755063194624,"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."}}