{"id":"W4399452441","doi":"10.2139/ssrn.4851778","title":"LOLA: LLM-Assisted Online Learning Algorithm for Content Experiments","year":2024,"lang":"en","type":"preprint","venue":"SSRN Electronic Journal","topic":"Advanced Bandit Algorithms Research","field":"Decision Sciences","cited_by":12,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Toronto","funders":"","keywords":"Computer science; Content (measure theory); Online learning; Algorithm; Artificial intelligence; Multimedia; Mathematics","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":[0.004056979,0.00136087,0.001517646,0.002079665,0.001062416,0.001924327,0.004365713,0.003812056,0.02630443],"category_scores_gemma":[0.02003132,0.0006386236,0.0008778751,0.001663789,0.001014905,0.003353217,0.003863099,0.003032351,0.01347341],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001353615,"about_ca_system_score_gemma":0.002736038,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002208989,"about_ca_topic_score_gemma":0.00380963,"domain_scores_codex":[0.9971521,0.001400934,0.000153778,0.0004918284,0.000593783,0.0002075854],"domain_scores_gemma":[0.9932191,0.003595102,0.0003293234,0.001548693,0.0009455996,0.0003623532],"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":[0.001486186,0.0006718102,0.001258512,0.0003123929,0.0001231074,0.0001150763,0.0000902129,0.07223385,0.00924267,0.01900977,0.03665599,0.8588005],"study_design_scores_gemma":[0.0001901719,0.0001409758,0.0001951864,0.00001847298,0.00001823813,0.00004247981,0.00002026468,0.970692,0.005210064,0.01905388,0.004398053,0.0000201311],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.004328441,0.0001438674,0.9730945,0.0002841476,0.0001200113,0.0001861795,0.0004491324,0.01957271,0.001820954],"genre_scores_gemma":[0.1050133,0.00006671371,0.8845704,0.0004608705,0.0001351965,0.001010228,0.001445247,0.001273071,0.006024892],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.02630443,"threshold_uncertainty_score":0.08799708,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1884793554634404,"score_gpt":0.4652270319985441,"score_spread":0.2767476765351037,"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."}}