{"id":"W4415428234","doi":"10.3233/faia251306","title":"Assessing and Improving the Multilingual Visual Word Sense Disambiguation Ability of Vision-Language Models","year":2025,"lang":"en","type":"book-chapter","venue":"Frontiers in artificial intelligence and applications","topic":"Speech and dialogue systems","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Artificial Intelligence in Medicine (Canada)","funders":"","keywords":"Generalization; Task (project management); Set (abstract data type); Generative grammar; Lemma (botany); Word (group theory); Word-sense disambiguation; Generative model","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":[0.0005256594,0.0001855196,0.0002819077,0.0001807589,0.0001876629,0.0002311399,0.000274484,0.0001704403,0.00000210557],"category_scores_gemma":[0.00005994553,0.0001574627,0.00005787436,0.0001452775,0.0002822358,0.0002589669,0.0001963568,0.0002420268,0.000002093787],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00004690497,"about_ca_system_score_gemma":0.0001054495,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001479921,"about_ca_topic_score_gemma":0.00006383449,"domain_scores_codex":[0.9985806,0.00004296475,0.0005075783,0.0005151997,0.0001902816,0.0001634054],"domain_scores_gemma":[0.9989229,0.0002338964,0.0002351644,0.000458224,0.00009655913,0.00005325101],"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":[0.000007444427,0.00002908318,0.0000138004,0.00005548427,0.000009701602,0.000001137837,0.0008648253,0.0001845353,0.0003370383,0.1002086,0.00005039539,0.8982379],"study_design_scores_gemma":[0.00004848642,0.00004131824,0.0001025641,0.0002664315,0.00003895371,0.000004099129,0.002346633,0.5850469,0.002640745,0.4082786,0.0008336366,0.0003515954],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.00153939,0.001486946,0.990738,0.0001761155,0.0002638029,0.0007227252,0.00001509874,0.00003676381,0.00502112],"genre_scores_gemma":[0.9652761,0.0002470828,0.03219037,0.0000776989,0.0002051652,0.0001178967,0.00002911931,0.00001781801,0.001838736],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9637367,"threshold_uncertainty_score":0.6421142,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03518871009626741,"score_gpt":0.3228200114855175,"score_spread":0.2876313013892501,"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."}}