{"id":"W7106793816","doi":"10.48448/y312-p752","title":"Deploying Tiny LVLM Judges for Real-World Evaluation of Chart Models: Lessons Learned and Best Practices","year":2025,"lang":"","type":"other","venue":"Open MIND","topic":"","field":"","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta; York University","funders":"","keywords":"Chart; Limiting; Best practice; Robustness (evolution); Comprehension; Control chart","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":[0.01283139,0.003430303,0.001544382,0.001634114,0.001175732,0.004751009,0.006215767,0.003156058,0.01505856],"category_scores_gemma":[0.06050083,0.001005604,0.001379456,0.001283999,0.001816381,0.006463971,0.003699368,0.005016835,0.008754917],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.003592846,"about_ca_system_score_gemma":0.003745353,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.02070009,"about_ca_topic_score_gemma":0.03112701,"domain_scores_codex":[0.9895287,0.00483078,0.0006311864,0.002460436,0.002022279,0.0005264775],"domain_scores_gemma":[0.9792373,0.01057688,0.0005070665,0.004788617,0.003696089,0.001194049],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.002103838,0.001046349,0.007726102,0.001821713,0.0005031459,0.0007302105,0.001311919,0.1825175,0.02438191,0.01474346,0.2500999,0.5130141],"study_design_scores_gemma":[0.0003764906,0.0003615039,0.001376117,0.0002485332,0.00007021227,0.0001373163,0.0005202484,0.927025,0.02015206,0.01934392,0.03022877,0.0001598819],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.1050151,0.005949261,0.5603619,0.005548032,0.002491264,0.001367714,0.009550684,0.2838115,0.02590451],"genre_scores_gemma":[0.4601895,0.0009076506,0.4905235,0.002321211,0.0002651441,0.0008975134,0.02471611,0.01279489,0.007384522],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.02070009,"threshold_uncertainty_score":0.06785971,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.5494438296058353,"score_gpt":0.5221059307475344,"score_spread":0.02733789885830085,"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."}}