{"id":"W4389518777","doi":"10.18653/v1/2023.emnlp-main.257","title":"Transductive Learning for Textual Few-Shot Classification in API-based Embedding Models","year":2023,"lang":"en","type":"article","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":7,"is_retracted":false,"has_abstract":true,"ca_institutions":"École de Technologie Supérieure","funders":"Grand Équipement National De Calcul Intensif","keywords":"Computer science; Embedding; Artificial intelligence; Natural language processing; Shot (pellet); Machine learning; Chemistry","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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0004823872,0.00009526652,0.0001180517,0.0002285157,0.0000969261,0.00007868215,0.0003654477,0.00006024755,0.000006773362],"category_scores_gemma":[0.00004784291,0.00009525004,0.00005284286,0.0004994709,0.00001383421,0.0004518018,0.00003572392,0.0001341999,0.00001889423],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00006535099,"about_ca_system_score_gemma":0.00007159803,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00003022955,"about_ca_topic_score_gemma":0.00002389333,"domain_scores_codex":[0.9988598,0.00005005817,0.0002161879,0.0004178299,0.0001740969,0.0002819944],"domain_scores_gemma":[0.9994187,0.0002011201,0.00004204528,0.0002421185,0.00005343731,0.00004256022],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.000006207783,0.00001564072,0.0001011342,0.00001392274,0.00000307915,0.000001271151,0.001538252,0.8577746,0.001977812,0.1083126,0.00005230628,0.03020322],"study_design_scores_gemma":[0.0003616938,0.00002676528,0.0002361198,0.00001528135,0.00000158593,4.500546e-7,0.0005137972,0.9905005,0.0008576845,0.007186248,0.0001808763,0.0001190447],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.03753693,0.000007275125,0.9581872,0.001264085,0.00009857128,0.0002548507,6.129412e-7,0.0003858946,0.002264549],"genre_scores_gemma":[0.9278626,0.000001921507,0.07142722,0.00009258911,0.00003833329,0.000086663,0.000008091484,0.000009781626,0.0004727974],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.8903257,"threshold_uncertainty_score":0.3884184,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1290711266498333,"score_gpt":0.3314435657339576,"score_spread":0.2023724390841243,"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."}}