{"id":"W4416366377","doi":"10.1109/tvcg.2025.3634791","title":"Probing the Visualization Literacy of Vision Language Models: The Good, the Bad, and the Ugly","year":2025,"lang":"en","type":"article","venue":"IEEE Transactions on Visualization and Computer Graphics","topic":"Multimodal Machine Learning Applications","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"","keywords":"Visualization; Chart; Correctness; Comprehension; Data visualization; Program comprehension; Code (set theory); Visual reasoning; Key (lock)","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":[0.0007981639,0.0001704866,0.0001516804,0.0001592969,0.001191054,0.00043431,0.0006530877,0.00006622338,0.000001917786],"category_scores_gemma":[0.000009186262,0.00008063205,0.00008457068,0.001278667,0.0003587107,0.0003094086,0.000029361,0.0002849892,0.000001036905],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00001133454,"about_ca_system_score_gemma":0.00003715912,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001943741,"about_ca_topic_score_gemma":0.00003495682,"domain_scores_codex":[0.9983448,0.0005750225,0.0003445422,0.0003136758,0.0002779475,0.0001440471],"domain_scores_gemma":[0.9980806,0.0009025614,0.0001664489,0.0006514559,0.0001700337,0.00002896559],"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":[0.0000175026,0.00007038458,0.00003593562,0.00002251654,0.00003928243,1.045834e-7,0.009935677,0.01135668,0.00001858305,0.9550244,0.00005117866,0.02342779],"study_design_scores_gemma":[0.0005638955,0.0000451613,0.001162793,0.0000649812,0.00004358041,0.000005876049,0.0001650814,0.9906718,0.0001802159,0.006450289,0.0005579283,0.00008846352],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01334486,0.0002914429,0.9820858,0.003128811,0.0002074233,0.0007218189,0.000003750525,0.0001078876,0.0001082373],"genre_scores_gemma":[0.9967064,0.000304359,0.0004818607,0.002290703,0.00002813444,0.0001032813,0.000003572314,0.00001154812,0.00007017766],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9833615,"threshold_uncertainty_score":0.9160741,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.009481657295951497,"score_gpt":0.3001178442489346,"score_spread":0.290636186952983,"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."}}