{"id":"W4402754270","doi":"10.1109/cvpr52733.2024.02674","title":"LLM4SGG: Large Language Models for Weakly Supervised Scene Graph Generation","year":2024,"lang":"en","type":"article","venue":"","topic":"Multimodal Machine Learning Applications","field":"Computer Science","cited_by":26,"is_retracted":false,"has_abstract":true,"ca_institutions":"Kootenay Association for Science & Technology","funders":"","keywords":"Computer science; Artificial intelligence; Natural language processing; Graph; Language model; Theoretical computer science","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001898621,0.002342422,0.001409005,0.001161469,0.0009387973,0.001606236,0.00582004,0.002759051,0.008392759],"category_scores_gemma":[0.005896623,0.001226119,0.002750116,0.0008808513,0.001204936,0.003135138,0.003331287,0.004376017,0.004292372],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002265139,"about_ca_system_score_gemma":0.001909424,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01520978,"about_ca_topic_score_gemma":0.03062129,"domain_scores_codex":[0.9987457,0.0004754596,0.00005507754,0.0004436272,0.0001797051,0.0001004356],"domain_scores_gemma":[0.9981198,0.0009707123,0.0000948256,0.0004610204,0.0002446591,0.0001090094],"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":[0.0005761029,0.0004406139,0.002154843,0.0004587133,0.000250174,0.0005577196,0.0005669969,0.4926313,0.01105931,0.02656261,0.05914394,0.4055978],"study_design_scores_gemma":[0.0000193427,0.00001501835,0.00005812605,0.000005873362,0.000006313613,0.00001796687,0.00001267791,0.9878,0.001021602,0.008808224,0.002226475,0.000008369291],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.008552405,0.000356948,0.9563112,0.0005285643,0.000122043,0.0003035237,0.001778375,0.03040858,0.001638371],"genre_scores_gemma":[0.2023904,0.0002814686,0.7699733,0.001330324,0.0001358976,0.00127411,0.01350637,0.003294279,0.007813836],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01520978,"threshold_uncertainty_score":0.03024244,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02582463357507845,"score_gpt":0.3026615849444719,"score_spread":0.2768369513693935,"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."}}