{"id":"W4391630552","doi":"10.31234/osf.io/upn8e","title":"Using Compositionality to Learn Many Categories from Few Examples","year":2024,"lang":"en","type":"preprint","venue":"","topic":"Semantic Web and Ontologies","field":"Computer Science","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Natural Sciences and Engineering Research Council of Canada; NOMIS Stiftung","keywords":"Principle of compositionality; Computer science; Natural language processing; Linguistics; Cognitive science; Artificial intelligence; Psychology; Philosophy","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.002741613,0.0006381646,0.0004254384,0.0006121029,0.0006419793,0.001563276,0.001207022,0.00105186,0.005397185],"category_scores_gemma":[0.02046825,0.0004043296,0.0008575222,0.0003749729,0.002262149,0.007748854,0.002729377,0.001994435,0.0008157227],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0004724116,"about_ca_system_score_gemma":0.0006010543,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0009451408,"about_ca_topic_score_gemma":0.001635555,"domain_scores_codex":[0.9983198,0.0007533114,0.00006869489,0.0005522105,0.0002371245,0.00006890589],"domain_scores_gemma":[0.9926835,0.004965365,0.0004564684,0.001308738,0.0003567284,0.0002291947],"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":[0.0006438172,0.0004655688,0.01869249,0.001286608,0.0002878247,0.000388998,0.005633117,0.04573177,0.05658092,0.2486367,0.005638137,0.6160141],"study_design_scores_gemma":[0.00006145886,0.0002599845,0.004107053,0.0001336425,0.00008983813,0.0004982544,0.0008868271,0.2674073,0.02131487,0.6900445,0.01510543,0.00009076416],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1922795,0.0004115145,0.7892374,0.00135151,0.00007878776,0.000165973,0.0002501025,0.001172506,0.01505272],"genre_scores_gemma":[0.7168598,0.0002401505,0.2782596,0.0004198926,0.00003504684,0.0001504194,0.000545481,0.0001514053,0.003338229],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.005397185,"threshold_uncertainty_score":0.01805532,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1150456316956632,"score_gpt":0.3335866663014433,"score_spread":0.2185410346057801,"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."}}