{"id":"W4385194783","doi":"10.1109/tai.2023.3298588","title":"A Visual and Textual Information Fusion-Based Zero-Shot Framework for Hazardous Material Placard Detection and Recognition","year":2023,"lang":"en","type":"article","venue":"IEEE Transactions on Artificial Intelligence","topic":"Infrastructure Maintenance and Monitoring","field":"Engineering","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"Okanagan University College; University of British Columbia, Okanagan Campus; University of British Columbia","funders":"Mitacs","keywords":"Computer science; Artificial intelligence; Margin (machine learning); Fuse (electrical); Logarithm; Shot (pellet); Machine learning; Object (grammar); Object detection; Pattern recognition (psychology); Engineering; Mathematics","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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0001298687,0.0001550121,0.0001244694,0.0002573912,0.0002469722,0.0001237081,0.00004045464,0.0001593589,0.00002317679],"category_scores_gemma":[0.00002373677,0.0001631323,0.00004324866,0.0002474382,0.00005434975,0.0002548324,0.000001074532,0.0001889524,0.00004500749],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0000497174,"about_ca_system_score_gemma":0.00001366588,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00002453652,"about_ca_topic_score_gemma":0.00003552856,"domain_scores_codex":[0.9992058,0.00001357203,0.0002747797,0.0001504565,0.0001186951,0.0002366793],"domain_scores_gemma":[0.9996065,0.0001526126,0.00003336884,0.00007855767,0.00006658251,0.00006234393],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.0002910955,0.00001151005,0.000001217366,0.00008806267,0.00001702196,0.00000111226,0.0005500304,0.07560491,0.03100391,0.00008792598,0.00001378683,0.8923294],"study_design_scores_gemma":[0.0000617686,0.0001933223,0.00004235089,0.00008442858,0.00002560631,0.000005381223,0.0005440792,0.2785736,0.7133379,0.006779536,0.0001464278,0.0002056132],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.3251002,0.000002325978,0.6727937,0.00001965558,0.001480708,0.0002380719,0.00007287717,0.0002741686,0.00001830063],"genre_scores_gemma":[0.9973399,0.00004352149,0.00222625,0.00004110657,0.0001711619,0.0001351987,0.00001833223,0.00002107614,0.00000347711],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.8921238,"threshold_uncertainty_score":0.6652343,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02902754123884744,"score_gpt":0.2719300137416246,"score_spread":0.2429024725027771,"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."}}