{"id":"W4396831993","doi":"10.1145/3613904.3641960","title":"Human-LLM Collaborative Annotation Through Effective Verification of LLM Labels","year":2024,"lang":"en","type":"article","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":103,"is_retracted":false,"has_abstract":true,"ca_institutions":"Simon Fraser University","funders":"","keywords":"Annotation; Leverage (statistics); Computer science; Domain (mathematical analysis); Natural language processing; Reliability (semiconductor); Artificial intelligence; Information retrieval; Data science","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":["metaresearch"],"consensus_categories":[],"category_scores_codex":[0.06115983,0.003270684,0.002507101,0.006327831,0.004012008,0.006390476,0.00644663,0.00535677,0.006224176],"category_scores_gemma":[0.1895328,0.001376415,0.002636109,0.003233219,0.004298394,0.009495041,0.01585355,0.005383621,0.004303953],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.003708792,"about_ca_system_score_gemma":0.00755698,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.006371085,"about_ca_topic_score_gemma":0.01139829,"domain_scores_codex":[0.9082809,0.0605997,0.003488628,0.01458694,0.01185546,0.001188431],"domain_scores_gemma":[0.6962627,0.1891474,0.0161057,0.07147247,0.02459481,0.002416975],"domain_codex":null,"domain_gemma":"methods","domain_candidate":"methods","domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.001820764,0.0009397337,0.01936122,0.002557142,0.0008779435,0.001545018,0.02118985,0.04147886,0.05123562,0.04971904,0.04291329,0.7663614],"study_design_scores_gemma":[0.0003500498,0.0004086131,0.007202639,0.0008677591,0.0005061137,0.0007722013,0.004768414,0.6312634,0.07745922,0.1774118,0.09836171,0.0006279985],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.02455704,0.0006481488,0.9559777,0.002714363,0.0002465535,0.0006359848,0.0009340029,0.009563405,0.004722826],"genre_scores_gemma":[0.2699222,0.000234365,0.7203398,0.0009074833,0.0002097599,0.0007880019,0.002083711,0.001629798,0.003884846],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9388402,"threshold_uncertainty_score":0.323448,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01972640779648407,"score_gpt":0.3134079163418643,"score_spread":0.2936815085453802,"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."}}