{"id":"W4416034102","doi":"10.18653/v1/2025.findings-emnlp.838","title":"LLMs for Bayesian Optimization in Scientific Domains: Are We There Yet?","year":2025,"lang":"","type":"article","venue":"","topic":"Advanced Multi-Objective Optimization Algorithms","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Natural Sciences and Engineering Research Council of Canada; Canadian Institute for Advanced Research","keywords":"Bayesian probability; Key (lock); Bayesian inference; Bayesian optimization","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow","scholarly_communication"],"consensus_categories":[],"category_scores_codex":[0.0008854172,0.0005307712,0.0005667473,0.001215435,0.0008820332,0.001350934,0.001443351,0.000298688,0.000260212],"category_scores_gemma":[0.0004843482,0.0005475592,0.0002036041,0.005125032,0.000415474,0.001853022,0.0005512261,0.0003029167,0.00002427788],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0005944383,"about_ca_system_score_gemma":0.0006762759,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001873522,"about_ca_topic_score_gemma":0.0002150144,"domain_scores_codex":[0.9954658,0.0002284466,0.0009768744,0.001925728,0.0004896712,0.000913481],"domain_scores_gemma":[0.9966817,0.0004242847,0.0004406257,0.001313058,0.0009592903,0.0001809821],"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":[0.0000392064,0.0003842474,0.0003134424,0.0001131508,0.00003388396,0.00000909792,0.0007751104,0.9198043,0.0000344901,0.03008813,0.0009994445,0.04740545],"study_design_scores_gemma":[0.00270581,0.00007351342,0.0002451972,0.0005400645,0.0000209419,0.000002648384,0.0008466615,0.9820394,0.0007815076,0.008386645,0.003815129,0.0005425132],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.00003392121,0.001251354,0.9816096,0.005371334,0.004085603,0.002500933,0.00003669971,0.0002228915,0.004887694],"genre_scores_gemma":[0.03002046,0.0004208175,0.942543,0.0005677672,0.00006419358,0.0002901759,0.00002798877,0.00004951976,0.02601603],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.06223503,"threshold_uncertainty_score":0.9996976,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01814474607173039,"score_gpt":0.2931597976741683,"score_spread":0.2750150516024379,"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."}}