{"id":"W4415305391","doi":"10.1016/j.nlp.2025.100185","title":"OPT2CODE: A retrieval-augmented framework for solving linear programming problems","year":2025,"lang":"en","type":"article","venue":"Natural Language Processing Journal","topic":"Optimization and Mathematical Programming","field":"Engineering","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"Queen's University","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Executable; Solver; Code generation; Domain (mathematical analysis); Benchmark (surveying); Code (set theory); Linear programming; Integer programming","routes":{"ca_aff":true,"ca_fund":true,"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":[0.0004283884,0.0002134801,0.0002533871,0.0001930803,0.0003563937,0.000460002,0.0002264279,0.0001683462,0.00002573987],"category_scores_gemma":[0.0007902798,0.0001779735,0.0001226807,0.0005479733,0.00003764365,0.0003153874,0.0000350575,0.0008261251,0.000003173755],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001179264,"about_ca_system_score_gemma":0.0000658723,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":4.368078e-7,"about_ca_topic_score_gemma":7.991341e-7,"domain_scores_codex":[0.9986677,0.00001422556,0.0004408548,0.000176712,0.0002293523,0.0004711137],"domain_scores_gemma":[0.9993235,0.0001259818,0.0001079865,0.0001166822,0.0002193796,0.0001064205],"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":[0.00007517191,0.00009885621,0.00002754468,0.002888281,0.0001576428,0.00002066295,0.003378899,0.009058055,0.004587655,0.001930239,0.000482888,0.9772941],"study_design_scores_gemma":[0.001041619,0.00005410292,0.000007850902,0.002869154,0.0001085154,0.00008691769,0.001295987,0.9795567,0.004723613,0.003818013,0.006006622,0.0004308929],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.002777645,0.02464787,0.9699555,0.0003683035,0.0007035378,0.0004818196,0.000001967881,0.0006302041,0.0004331566],"genre_scores_gemma":[0.4896357,0.00004080347,0.5094289,0.0001315393,0.0002447327,0.00002111642,0.000008776889,0.00004321531,0.0004452203],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.9768632,"threshold_uncertainty_score":0.7257548,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.00951167119287442,"score_gpt":0.2937268453066785,"score_spread":0.284215174113804,"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."}}