{"id":"W7077911890","doi":"10.48448/7ydn-np56","title":"ResearchAgent: Iterative Research Idea Generation over Scientific Literature with Large Language Models","year":2025,"lang":"en","type":"other","venue":"Underline Science Inc.","topic":"Topic Modeling","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Kootenay Association for Science & Technology","funders":"","keywords":"Operationalization; Leverage (statistics); Pace; Language model; Scientific literature; Readability; Natural language generation","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":[0.01755203,0.002200178,0.001256346,0.003515629,0.001139805,0.004633304,0.004055016,0.002319296,0.007430641],"category_scores_gemma":[0.06767564,0.001321966,0.002269945,0.001653985,0.001401977,0.008203057,0.007164414,0.002194605,0.004665955],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001332676,"about_ca_system_score_gemma":0.003017131,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00188344,"about_ca_topic_score_gemma":0.004196798,"domain_scores_codex":[0.9876881,0.007799333,0.0008343005,0.001700756,0.001770268,0.0002071917],"domain_scores_gemma":[0.9314007,0.04966849,0.003349702,0.009381078,0.004574236,0.001625712],"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":[0.00196085,0.001508754,0.01123626,0.002947437,0.000926972,0.00109486,0.007937774,0.05039306,0.03730615,0.04340507,0.0667607,0.774522],"study_design_scores_gemma":[0.0006960353,0.0006232221,0.001355732,0.0002533059,0.0003422895,0.0004114876,0.001129667,0.8302452,0.02496384,0.05818455,0.08156812,0.0002266153],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.02161478,0.0005467556,0.9021878,0.001202452,0.0002169269,0.001243249,0.0008482898,0.06870091,0.003438953],"genre_scores_gemma":[0.1189908,0.0002882321,0.8707131,0.0005127469,0.0001398969,0.001256637,0.00247476,0.001929672,0.003694123],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.01755203,"threshold_uncertainty_score":0.09282511,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.06560478267328029,"score_gpt":0.3721665830719578,"score_spread":0.3065618003986775,"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."}}