{"id":"W4385688511","doi":"10.1145/3578337.3605136","title":"Perspectives on Large Language Models for Relevance Judgment","year":2023,"lang":"en","type":"article","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":152,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"National Science Foundation","keywords":"Relevance (law); Perspective (graphical); Categorization; Computer science; Point (geometry); Compromise; Cognitive psychology; Psychology; Artificial intelligence; Sociology; Political science; Social 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":[],"consensus_categories":[],"category_scores_codex":[0.07345315,0.001647202,0.002302094,0.004681685,0.002729954,0.01431539,0.004907787,0.005755314,0.006370403],"category_scores_gemma":[0.1661479,0.001530505,0.001802058,0.002982647,0.01235336,0.02932009,0.007264181,0.009937408,0.002138304],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.004974428,"about_ca_system_score_gemma":0.003059765,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003921557,"about_ca_topic_score_gemma":0.002970466,"domain_scores_codex":[0.9459254,0.04274909,0.001503309,0.003890956,0.005073111,0.0008580969],"domain_scores_gemma":[0.6764201,0.2811487,0.00691776,0.02495913,0.007913221,0.002641122],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.0001170609,0.00007078386,0.0009936986,0.0003161844,0.0001113057,0.00008174834,0.002594724,0.009490547,0.0008249442,0.9440781,0.0039639,0.03735708],"study_design_scores_gemma":[0.00003135169,0.00002739074,0.0003584649,0.00008109859,0.00002287249,0.00005682647,0.0002719826,0.04835247,0.0004602701,0.9432694,0.007025952,0.00004186442],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":"theoretical_or_conceptual","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01547918,0.009073187,0.8866804,0.06047148,0.0003983487,0.0001458004,0.0003051957,0.001379794,0.02606669],"genre_scores_gemma":[0.6278574,0.00330703,0.3549328,0.004682542,0.002927525,0.0008475192,0.0004982935,0.0006766091,0.004270291],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.07345315,"threshold_uncertainty_score":0.388462,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03503074496303086,"score_gpt":0.2930385171724492,"score_spread":0.2580077722094183,"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."}}