{"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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0002828671,0.0001103712,0.00009211145,0.0001303625,0.0001446055,0.0003171907,0.0004413209,0.00004994637,0.00005822487],"category_scores_gemma":[0.00001918546,0.00009624018,0.00008625413,0.0003803263,0.000009406353,0.0004383921,0.00009959273,0.0001053805,0.0001089248],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0000248117,"about_ca_system_score_gemma":0.00004617449,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0002296967,"about_ca_topic_score_gemma":0.00005168842,"domain_scores_codex":[0.9990013,0.00003170663,0.0001565176,0.0004268029,0.000152841,0.0002308167],"domain_scores_gemma":[0.9993168,0.00008178197,0.00001676438,0.0004637807,0.00005368808,0.00006721741],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.000001123817,0.00004612381,0.00002560257,0.00003547468,0.00001644197,0.000002047487,0.001153579,0.005744387,0.02731567,0.9194241,0.003857806,0.04237763],"study_design_scores_gemma":[0.0001463135,0.00002096628,0.00009291025,0.000007797878,0.000005778144,0.000002753707,0.00001832227,0.9878423,0.002427757,0.006879673,0.002430173,0.0001253079],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01625936,0.0003609657,0.9766796,0.002630693,0.0002052591,0.0003588368,0.00001741518,0.0008519568,0.002635972],"genre_scores_gemma":[0.7414111,0.000005965071,0.2569471,0.0003997954,0.0001507552,0.0001612562,0.00004728977,0.00001455506,0.0008621598],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9820979,"threshold_uncertainty_score":0.3924561,"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."}}