{"id":"W4415269798","doi":"10.48550/arxiv.2510.12141","title":"MAPS: Masked Attribution-based Probing of Strategies- A computational framework to align human and model explanations","year":2025,"lang":"en","type":"preprint","venue":"ArXiv.org","topic":"Explainable Artificial Intelligence (XAI)","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Natural Sciences and Engineering Research Council of Canada; Canada First Research Excellence Fund","keywords":"Interpretability; Set (abstract data type); Computational model; Similarity (geometry); Psychophysics; Attribution; Object (grammar); Artificial neural network","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0004444309,0.0002993262,0.0003845531,0.000407287,0.0003077423,0.0002550333,0.001081008,0.0002713741,0.00001158566],"category_scores_gemma":[0.0001465994,0.0003401371,0.000108041,0.0005997262,0.0001039573,0.0003126643,0.0008629067,0.0004797255,0.00002084693],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001569607,"about_ca_system_score_gemma":0.0009632234,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0002077176,"about_ca_topic_score_gemma":0.00006853634,"domain_scores_codex":[0.997681,0.0001123891,0.0006403027,0.0008180953,0.0003929559,0.0003552372],"domain_scores_gemma":[0.9977862,0.0004637175,0.0002739106,0.0007964662,0.0005477304,0.000131964],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.000004582759,0.00006867246,0.003171267,0.0001127874,0.00002822797,0.00000376313,0.001189499,0.6627167,0.0002843254,0.3321223,0.0001433,0.0001545632],"study_design_scores_gemma":[0.00008563581,0.00005536622,0.00211443,0.0005651232,0.00002058977,5.777852e-7,0.0002332807,0.685511,0.003863042,0.3071662,0.00005072513,0.0003339652],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1969856,0.00004331551,0.8001273,0.001518691,0.0001785852,0.0005691691,0.00008657023,0.0001420733,0.0003487119],"genre_scores_gemma":[0.8687176,0.000003075614,0.1304478,0.0004106485,0.00004101361,0.0001569798,0.0001079367,0.00001135825,0.0001035675],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.6717319,"threshold_uncertainty_score":0.999905,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.09027916531824774,"score_gpt":0.3470289235041251,"score_spread":0.2567497581858773,"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."}}