{"id":"W4416037063","doi":"10.18653/v1/2025.emnlp-main.136","title":"SurveyGen: Quality-Aware Scientific Survey Generation with Large Language Models","year":2025,"lang":"","type":"article","venue":"","topic":"Survey Methodology and Nonresponse","field":"Social Sciences","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Alliance de recherche numérique du Canada; Natural Sciences and Engineering Research Council of Canada; National Natural Science Foundation of China; University of Alberta","keywords":"Natural language generation; Natural language; Language model; Government (linguistics)","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.01417349,0.001584186,0.001039557,0.006370005,0.0008276563,0.002746325,0.002306178,0.001640507,0.008016452],"category_scores_gemma":[0.06845123,0.0007566214,0.001648102,0.004148487,0.0007084713,0.003232378,0.003995595,0.001638229,0.00892555],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001092201,"about_ca_system_score_gemma":0.00532219,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003472591,"about_ca_topic_score_gemma":0.009883648,"domain_scores_codex":[0.9887844,0.006833085,0.001203362,0.001501041,0.001418012,0.0002600059],"domain_scores_gemma":[0.9478775,0.03220899,0.002822256,0.01078459,0.005236057,0.001070533],"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":[0.000752318,0.0006123696,0.02462635,0.005195199,0.0004140366,0.0003367147,0.002284549,0.01429113,0.016658,0.01028438,0.3471422,0.5774027],"study_design_scores_gemma":[0.001379038,0.0008848621,0.01819212,0.0008207811,0.0003212475,0.000799157,0.002149662,0.448332,0.04120085,0.05885835,0.4266297,0.000432254],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.04252321,0.003291928,0.6053602,0.003139441,0.0008133097,0.003704698,0.1247597,0.2093164,0.007091149],"genre_scores_gemma":[0.1086538,0.0007989076,0.6851587,0.001328223,0.0002168667,0.004270832,0.1921004,0.003191256,0.004281],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.01417349,"threshold_uncertainty_score":0.07495749,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.3905043516417337,"score_gpt":0.498332670281995,"score_spread":0.1078283186402612,"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."}}