{"id":"W4320558549","doi":"10.48550/arxiv.2302.04983","title":"CREDENCE: Counterfactual Explanations for Document Ranking","year":2023,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Advanced Text Analysis Techniques","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Toronto Metropolitan University; York University; University of Waterloo","funders":"","keywords":"Credence; Counterfactual thinking; Ranking (information retrieval); Information retrieval; Computer science; Attribution; Machine learning; Psychology; Social psychology","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":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0003570743,0.0002911415,0.0003408392,0.0004849089,0.0002565501,0.0002101761,0.002255973,0.0002141673,0.00001877117],"category_scores_gemma":[0.00007793275,0.0003483537,0.0002891858,0.0006279889,0.0000726729,0.0006138856,0.001742286,0.0003511559,0.00006510982],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0003955946,"about_ca_system_score_gemma":0.0001579332,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00008146151,"about_ca_topic_score_gemma":0.00008364689,"domain_scores_codex":[0.9980326,0.00006234776,0.0002529079,0.001139766,0.0001456529,0.0003667031],"domain_scores_gemma":[0.9978176,0.0003067198,0.0002989747,0.001221023,0.0002528327,0.0001027948],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.00002396732,0.00006546345,0.0003552738,0.00009243406,0.0002790654,0.0001252915,0.0005890675,0.2481926,0.00006595143,0.7445979,0.003748276,0.001864671],"study_design_scores_gemma":[0.000378739,0.00005493027,0.0001317955,0.0001305022,0.0001211426,0.000001856811,0.0001383909,0.3938088,0.0006568627,0.6004897,0.00349935,0.0005879405],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":"theoretical_or_conceptual","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.009263458,0.00002777222,0.9880554,0.0002305405,0.000404411,0.0005042445,0.00002781608,0.0011353,0.0003510245],"genre_scores_gemma":[0.9599088,0.0001916008,0.03636212,0.00007519893,0.00007260532,0.00001504678,0.00005006347,0.00002676977,0.003297837],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9516933,"threshold_uncertainty_score":0.9998968,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1383141408168375,"score_gpt":0.2483798991809904,"score_spread":0.1100657583641529,"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."}}