{"id":"W7126433153","doi":"10.21428/594757db.4bc3ecf6","title":"Instruction Tuning of LLMs for Multi-label EmotionClassification in Social Media Content","year":2024,"lang":"en","type":"article","venue":"","topic":"Sentiment Analysis and Opinion Mining","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Ottawa","funders":"","keywords":"Controllability; Inference; Social media; Generative model; Test (biology); Order (exchange); Training set; Language model","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":[0.001517203,0.001644295,0.0007653131,0.000734534,0.0004118654,0.001175353,0.001844753,0.001293736,0.005157932],"category_scores_gemma":[0.009644864,0.0004811062,0.0008562722,0.0004274525,0.0005578689,0.002932629,0.001497718,0.003053134,0.00323384],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000992695,"about_ca_system_score_gemma":0.00117064,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003412057,"about_ca_topic_score_gemma":0.005293545,"domain_scores_codex":[0.9990577,0.0002781373,0.00005078519,0.0003962118,0.000121579,0.00009563624],"domain_scores_gemma":[0.9976758,0.001334856,0.00009970431,0.0003689628,0.0003951389,0.0001255765],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"not_applicable","study_design_scores_codex":[0.001248536,0.0009924198,0.0102498,0.0004233389,0.0001599485,0.0003616609,0.0007029814,0.1363396,0.057958,0.004162773,0.01870165,0.7686993],"study_design_scores_gemma":[0.00007097152,0.0001487096,0.0009848394,0.00001604509,0.00003218503,0.00006309466,0.000105575,0.9714471,0.0187072,0.005641378,0.002755426,0.00002749321],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1315995,0.0007626942,0.8004712,0.0008029777,0.0004651562,0.000370151,0.001126537,0.06160444,0.002797284],"genre_scores_gemma":[0.758441,0.0001713259,0.2310101,0.0007784946,0.0001429412,0.0005424953,0.003104293,0.002029046,0.003780417],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.005157932,"threshold_uncertainty_score":0.01725495,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2369569092403301,"score_gpt":0.3485233860888012,"score_spread":0.1115664768484711,"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."}}