{"id":"W4402963229","doi":"10.1145/3664647.3680815","title":"Text-Region Matching for Multi-Label Image Recognition with Missing Labels","year":2024,"lang":"en","type":"preprint","venue":"","topic":"Multimodal Machine Learning Applications","field":"Computer Science","cited_by":7,"is_retracted":false,"has_abstract":true,"ca_institutions":"Artificial Intelligence in Medicine (Canada)","funders":"RIKEN; Natural Science Foundation of Anhui Province; National Natural Science Foundation of China","keywords":"Computer science; Bridging (networking); Pascal (unit); Artificial intelligence; Matching (statistics); Semantics (computer science); Benchmark (surveying); Class (philosophy); Margin (machine learning); Pattern recognition (psychology); Image (mathematics); Visualization; Semantic gap; Natural language processing; Machine learning; Image retrieval; Mathematics","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":["metaepi_narrow","scholarly_communication"],"consensus_categories":[],"category_scores_codex":[0.0003588668,0.0003867387,0.0003130741,0.000246353,0.0002526327,0.001079198,0.000960915,0.0002257993,0.00001146218],"category_scores_gemma":[0.00006603081,0.0003250172,0.0001104818,0.0002761757,0.0000499357,0.0002283451,0.001327071,0.0009943336,0.0002689141],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001233082,"about_ca_system_score_gemma":0.0001920829,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0009098612,"about_ca_topic_score_gemma":0.00004446523,"domain_scores_codex":[0.9977344,0.00007661403,0.0003633728,0.001211123,0.000250926,0.0003635628],"domain_scores_gemma":[0.9982417,0.000217776,0.000240708,0.0009323639,0.0002420853,0.0001253349],"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":[0.00002245759,0.0003197763,0.00005693281,0.001390908,0.0001499207,0.00002590104,0.001800816,0.0009125012,0.006345236,0.01792407,0.001523121,0.9695284],"study_design_scores_gemma":[0.0005668666,0.00004749545,0.0003157558,0.0006886498,0.00006003453,0.00003836936,0.00002180436,0.9146438,0.001325294,0.08157394,0.0001852212,0.0005327895],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.01251256,0.0001063968,0.9746267,0.008361388,0.0002904796,0.001259121,0.00002523282,0.001385762,0.001432395],"genre_scores_gemma":[0.07049713,0.000007742447,0.9273075,0.0003295319,0.0001478372,0.0006023333,0.00009910433,0.00007270531,0.0009361075],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.9689956,"threshold_uncertainty_score":0.9999578,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.07659082094066458,"score_gpt":0.3466163301374486,"score_spread":0.270025509196784,"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."}}