{"id":"W4317384342","doi":"10.1109/access.2023.3237966","title":"A Cross-Modal Alignment for Zero-Shot Image Classification","year":2023,"lang":"en","type":"article","venue":"IEEE Access","topic":"Domain Adaptation and Few-Shot Learning","field":"Computer Science","cited_by":5,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Ministry of Natural Resources","keywords":"Computer science; Matching (statistics); Artificial intelligence; Feature (linguistics); Pattern recognition (psychology); Embedding; Metric (unit); Modal; Feature extraction; Zero (linguistics); Image (mathematics); Encoder; Contextual image classification; Measure (data warehouse); Key (lock); Data mining; Mathematics","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":[0.001331101,0.001143258,0.00146669,0.001648949,0.0008628042,0.001017027,0.002356325,0.001511424,0.004485541],"category_scores_gemma":[0.002938302,0.000402115,0.001063414,0.001750809,0.0008683926,0.002506549,0.002656207,0.002306517,0.002932334],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0006879062,"about_ca_system_score_gemma":0.001006826,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003972109,"about_ca_topic_score_gemma":0.005954661,"domain_scores_codex":[0.9982007,0.0002675489,0.00007198674,0.0009268685,0.0003280414,0.000204801],"domain_scores_gemma":[0.9988123,0.0002477077,0.00009674423,0.0004068608,0.000337863,0.00009841178],"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":[0.0005040472,0.0007205148,0.002336588,0.0001852022,0.0001194341,0.0001005634,0.000213933,0.0166755,0.05133945,0.004563583,0.01231766,0.9109235],"study_design_scores_gemma":[0.00004163587,0.0003529377,0.002910191,0.00002478147,0.00006237701,0.0003186968,0.0001550154,0.9376742,0.03999073,0.0116711,0.006739578,0.00005884693],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.03305349,0.0007269146,0.9541523,0.0001568065,0.0001661062,0.0001468094,0.0004417099,0.008654283,0.002501499],"genre_scores_gemma":[0.5073566,0.0004321077,0.4742062,0.0005903982,0.0002228517,0.0003964944,0.005704826,0.0007688365,0.01032164],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.004485541,"threshold_uncertainty_score":0.01500565,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1308000278195402,"score_gpt":0.4054642342831136,"score_spread":0.2746642064635734,"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."}}