{"id":"W7082289080","doi":"10.48448/ffmv-8j70","title":"MAmmoTH-VL: Eliciting Multimodal Reasoning with Instruction Tuning at Scale","year":2025,"lang":"en","type":"other","venue":"Underline Science Inc.","topic":"Geochemistry and Geologic Mapping","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"","keywords":"Construct (python library); Scalability; Scale (ratio); Range (aeronautics); Cover (algebra)","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":[0.00198091,0.00340163,0.000844634,0.001821684,0.0007793119,0.002049459,0.004158188,0.002858222,0.01643445],"category_scores_gemma":[0.01481171,0.0007057199,0.002084741,0.001633766,0.0009672471,0.004562147,0.003728559,0.003623906,0.00797139],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0019707,"about_ca_system_score_gemma":0.002676247,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01392917,"about_ca_topic_score_gemma":0.03375953,"domain_scores_codex":[0.9973112,0.0009274434,0.0001808917,0.001032343,0.0004018524,0.0001462435],"domain_scores_gemma":[0.9951313,0.002947985,0.0001761198,0.0011483,0.0004018428,0.0001945641],"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":[0.001317742,0.001477821,0.01049687,0.003775185,0.0004253447,0.0006650373,0.0007911856,0.08407024,0.0177559,0.01434505,0.3387327,0.5261469],"study_design_scores_gemma":[0.0006291595,0.0005438297,0.004520939,0.0003431367,0.0001777201,0.0004592346,0.0009625378,0.7753333,0.02707673,0.04905316,0.1407229,0.0001773387],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.1303574,0.006291328,0.4060496,0.004888363,0.001012054,0.001914385,0.1537977,0.2590706,0.03661854],"genre_scores_gemma":[0.2783009,0.0009379031,0.4478969,0.002297404,0.0001472669,0.001445387,0.2563393,0.002989234,0.009645639],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.01643445,"threshold_uncertainty_score":0.05497873,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.007702649910607454,"score_gpt":0.2303360047589924,"score_spread":0.222633354848385,"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."}}