{"id":"W4386566765","doi":"10.18653/v1/2023.eacl-main.185","title":"MAPL: Parameter-Efficient Adaptation of Unimodal Pre-Trained Models for Vision-Language Few-Shot Prompting","year":2023,"lang":"en","type":"article","venue":"","topic":"Multimodal Machine Learning Applications","field":"Computer Science","cited_by":19,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université de Montréal; Canadian Institute for Advanced Research","funders":"Samsung; Canadian Institute for Advanced Research","keywords":"Computer science; Adaptation (eye); Artificial intelligence; Computational linguistics; Association (psychology); Shot (pellet); Natural language processing; Speech recognition; Psychology; Neuroscience; Chemistry","routes":{"ca_aff":true,"ca_fund":true,"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":[0.001064927,0.001730235,0.001430336,0.0007305322,0.0004491185,0.001250076,0.003681131,0.001793715,0.01532026],"category_scores_gemma":[0.004701546,0.0007567443,0.0007753031,0.0005122645,0.0003890273,0.002069979,0.003014379,0.002875389,0.008577514],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0008021772,"about_ca_system_score_gemma":0.00103767,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.007768478,"about_ca_topic_score_gemma":0.01186581,"domain_scores_codex":[0.9994776,0.0001136759,0.0000222411,0.0002047941,0.0001103541,0.00007136937],"domain_scores_gemma":[0.9989709,0.0004579345,0.00003338502,0.0002034077,0.0002311319,0.0001033329],"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":[0.001187525,0.0004433092,0.0006016506,0.0003517416,0.000160239,0.0003760637,0.0002142718,0.077713,0.03225641,0.002969185,0.05420884,0.8295178],"study_design_scores_gemma":[0.0000613192,0.0001036578,0.0003013646,0.00002280029,0.00002383893,0.0001005042,0.00006024832,0.9741758,0.01487517,0.004261294,0.005982837,0.00003116183],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01477712,0.0009815102,0.8798054,0.0001980521,0.0005290943,0.0001955981,0.001048453,0.09925348,0.003211135],"genre_scores_gemma":[0.3305669,0.0006488507,0.6383891,0.0007183506,0.0002450581,0.0007566141,0.006255395,0.007068627,0.01535107],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01532026,"threshold_uncertainty_score":0.05125141,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04356935763959081,"score_gpt":0.3403577905972081,"score_spread":0.2967884329576173,"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."}}