{"id":"W4407737256","doi":"10.1109/ickg63256.2024.00017","title":"Axolotl: Fairness through Assisted Prompt Rewriting of Large Language Model Outputs","year":2024,"lang":"en","type":"article","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"National Science Foundation","keywords":"Axolotl; Rewriting; Computer science; Programming language; Regeneration (biology)","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":[],"consensus_categories":[],"category_scores_codex":[0.0003359157,0.0001165694,0.0001690855,0.00006541444,0.0000484918,0.0001237284,0.00056881,0.00006567199,0.00001847136],"category_scores_gemma":[0.00003309978,0.00009797273,0.0000764306,0.000285532,0.00001618319,0.0005645753,0.0003434175,0.0001385392,0.00002748372],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00002931258,"about_ca_system_score_gemma":0.00008517414,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00007617867,"about_ca_topic_score_gemma":0.0000244436,"domain_scores_codex":[0.9987047,0.00003283993,0.0002886354,0.0004189037,0.0002742314,0.0002807302],"domain_scores_gemma":[0.9992772,0.00005719731,0.00004090574,0.0005307936,0.0000549813,0.00003888676],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.000001424658,0.0000547263,0.00005806111,0.0002218119,0.00002380704,0.00005200516,0.007046703,0.001463831,0.004906156,0.9609346,0.0005205403,0.0247163],"study_design_scores_gemma":[0.0001219326,0.00001229362,0.00004579,0.0001095691,0.000005159369,0.00001265204,0.0002396369,0.9876623,0.0061939,0.005122492,0.0003479288,0.0001263489],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.03413366,0.0005078021,0.9443501,0.0007724377,0.0002050727,0.0001214773,0.000004487663,0.0006494071,0.01925555],"genre_scores_gemma":[0.8066284,0.000002779077,0.1912767,0.0001796098,0.00004696514,0.000008926318,0.000001777935,0.00001143491,0.001843407],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9861985,"threshold_uncertainty_score":0.3995212,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03888180439271441,"score_gpt":0.3118506534979791,"score_spread":0.2729688491052648,"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."}}