{"id":"W4412377079","doi":"10.1145/3726302.3730076","title":"Reverse-Engineering the Retrieval Process in GenIR Models","year":2025,"lang":"en","type":"article","venue":"","topic":"Advanced Image and Video Retrieval Techniques","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Azrieli Foundation; European Commission; Open Philanthropy Project","keywords":"Computer science; Process (computing); Reverse engineering; Information retrieval; Data science; Software engineering; Artificial intelligence; Programming language","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.00195684,0.0009645908,0.0008971312,0.0006331049,0.0003469376,0.001281788,0.001971829,0.001485276,0.00504567],"category_scores_gemma":[0.005632198,0.0006855877,0.001143023,0.0005664226,0.001073573,0.00318822,0.001313669,0.001995346,0.002558655],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0008241094,"about_ca_system_score_gemma":0.0007763503,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004232102,"about_ca_topic_score_gemma":0.004963967,"domain_scores_codex":[0.9993349,0.0002231807,0.00003687731,0.0002007925,0.0001213258,0.00008293392],"domain_scores_gemma":[0.998421,0.0009658559,0.0001132215,0.0002629272,0.0001919748,0.00004511416],"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":[0.0005651482,0.0002388118,0.002651573,0.0005433399,0.0002076458,0.0007148054,0.0007784384,0.5790767,0.03796493,0.1167753,0.004568988,0.2559143],"study_design_scores_gemma":[0.00001847476,0.00008776349,0.0002419427,0.00001328462,0.00004456953,0.0001800852,0.00002588397,0.9750455,0.004446003,0.01850873,0.001367816,0.00001997528],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.05144981,0.0009062056,0.93595,0.0006359112,0.00009225348,0.0001430088,0.0002154487,0.002045601,0.008561779],"genre_scores_gemma":[0.768903,0.001085478,0.2014879,0.0007137926,0.0001252906,0.0003309653,0.0004808054,0.0004971874,0.02637555],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.00504567,"threshold_uncertainty_score":0.0168795,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01114301822334733,"score_gpt":0.2771883083184644,"score_spread":0.266045290095117,"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."}}