{"id":"W4409263021","doi":"10.1109/wacv61041.2025.00265","title":"AiDe: Improving 3D Open-Vocabulary Semantic Segmentation by Aligned Vision-Language Learning","year":2025,"lang":"en","type":"article","venue":"","topic":"Multimodal Machine Learning Applications","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"","keywords":"Computer science; Natural language processing; Segmentation; Artificial intelligence; Vocabulary; Vocabulary learning; Image segmentation; Linguistics","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":[0.0004959332,0.0001639563,0.0001685345,0.0001481247,0.0004296108,0.0006884261,0.001396392,0.00006429199,0.00006856915],"category_scores_gemma":[0.0001339735,0.0001575343,0.00003936421,0.0006526917,0.00002632248,0.000708846,0.001013133,0.000199187,0.0001517554],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00007201015,"about_ca_system_score_gemma":0.00008007608,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003237567,"about_ca_topic_score_gemma":0.00002905983,"domain_scores_codex":[0.9984542,0.000169024,0.0002956547,0.0005924412,0.0002196111,0.0002691095],"domain_scores_gemma":[0.9989285,0.0002174415,0.0001224539,0.0006169991,0.00004226698,0.00007234996],"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":[0.000008894939,0.0001915581,0.006065048,0.00006322202,0.00004509646,0.000008165735,0.001465715,0.002924389,0.3164476,0.01268198,0.005355066,0.6547432],"study_design_scores_gemma":[0.0007103432,0.00007492647,0.005691735,0.00004515971,0.00001457871,0.000004361767,0.0002621739,0.9760489,0.01488444,0.0003568841,0.001639891,0.00026655],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.05490888,0.0001417454,0.9312276,0.002710847,0.0001139226,0.0004909675,8.536271e-7,0.0004846041,0.009920579],"genre_scores_gemma":[0.8267138,0.000004735589,0.1657822,0.001093022,0.00001583841,0.00008382595,0.00003184088,0.0000127211,0.006262043],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9731246,"threshold_uncertainty_score":0.6638508,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.006589663231590595,"score_gpt":0.3060305378835126,"score_spread":0.299440874651922,"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."}}