{"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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0007825024,0.002872565,0.00214709,0.001916766,0.0007022203,0.001909661,0.004070482,0.002303368,0.006690259],"category_scores_gemma":[0.002766819,0.0009668542,0.002518901,0.001680219,0.00127374,0.004813479,0.004103614,0.002438932,0.005796013],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001152814,"about_ca_system_score_gemma":0.00163433,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.008767419,"about_ca_topic_score_gemma":0.01319805,"domain_scores_codex":[0.99873,0.0001390452,0.00005746583,0.0005800805,0.000338702,0.0001546027],"domain_scores_gemma":[0.9992288,0.0001806569,0.00005515076,0.0002658556,0.0001951241,0.00007433721],"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":[0.0005146621,0.0004556215,0.00135847,0.0004878251,0.0002327886,0.0005107042,0.0004313197,0.1187249,0.06201575,0.01263905,0.03362378,0.7690052],"study_design_scores_gemma":[0.00006155825,0.0001500809,0.0004712902,0.00003315278,0.00004183401,0.000233137,0.0001422083,0.9434205,0.02727228,0.01794369,0.01017058,0.00005977411],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.02082024,0.0007712982,0.9430135,0.000173364,0.000219156,0.000185946,0.001700911,0.02952999,0.003585599],"genre_scores_gemma":[0.2429066,0.0005462373,0.7227834,0.001004313,0.0001376757,0.0005997009,0.01986588,0.002594804,0.009561257],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.008767419,"threshold_uncertainty_score":0.02238119,"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."}}