{"id":"W4401042217","doi":"10.18653/v1/2024.naacl-industry.33","title":"Tiny Titans: Can Smaller Large Language Models Punch Above Their Weight in the Real World for Meeting Summarization?","year":2024,"lang":"en","type":"article","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":17,"is_retracted":false,"has_abstract":true,"ca_institutions":"Stornoway Diamond (Canada)","funders":"","keywords":"Automatic summarization; Computer science; Artificial intelligence","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":[0.004812986,0.001747565,0.00172621,0.001497144,0.001455171,0.003736457,0.002816516,0.002136833,0.01697463],"category_scores_gemma":[0.03227606,0.00100805,0.0009789122,0.001436524,0.001064949,0.01580174,0.002635362,0.003142769,0.01477358],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0008821057,"about_ca_system_score_gemma":0.001258994,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00457449,"about_ca_topic_score_gemma":0.01206545,"domain_scores_codex":[0.9980845,0.0009576823,0.0001185675,0.0004047013,0.0003146492,0.0001197306],"domain_scores_gemma":[0.9910761,0.005368848,0.0003685561,0.001776451,0.001002347,0.0004075819],"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":[0.003073433,0.0002772807,0.003118521,0.0006694582,0.0003769103,0.0002223245,0.001315659,0.02367534,0.008131537,0.03413358,0.2055535,0.7194524],"study_design_scores_gemma":[0.0006791994,0.0005597133,0.001639064,0.0001959197,0.0002749948,0.0002440749,0.001341947,0.6641403,0.007846409,0.2147408,0.1081999,0.0001375964],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.06203199,0.0133221,0.8089259,0.03972761,0.006130704,0.0004128349,0.006028879,0.04194003,0.02147995],"genre_scores_gemma":[0.5111642,0.003881492,0.4391603,0.005834759,0.003332988,0.0006039257,0.01162703,0.005004846,0.01939038],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.01697463,"threshold_uncertainty_score":0.05678576,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02476747805012692,"score_gpt":0.2657033425659284,"score_spread":0.2409358645158015,"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."}}