{"id":"W4401042661","doi":"10.18653/v1/2024.naacl-long.208","title":"Long-form evaluation of model editing","year":2024,"lang":"en","type":"article","venue":"","topic":"Model-Driven Software Engineering Techniques","field":"Computer Science","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"Dalhousie University","funders":"Alliance de recherche numérique du Canada; Killam Trusts","keywords":"Computer science","routes":{"ca_aff":true,"ca_fund":true,"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":[0.01231104,0.001553307,0.001310576,0.001910349,0.00149483,0.003607577,0.002753351,0.002254578,0.01940781],"category_scores_gemma":[0.09330823,0.000457575,0.0009157503,0.001592633,0.001242538,0.00611356,0.004055414,0.001678805,0.005789365],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001483044,"about_ca_system_score_gemma":0.001894231,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004871128,"about_ca_topic_score_gemma":0.01013377,"domain_scores_codex":[0.9840556,0.007424149,0.0009956874,0.001846168,0.004978701,0.000699641],"domain_scores_gemma":[0.927007,0.03311859,0.001566177,0.01912219,0.01770885,0.001477076],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"not_applicable","study_design_scores_codex":[0.003571371,0.001083945,0.006997089,0.001043093,0.0004961729,0.0007816286,0.0009362894,0.1375841,0.009488152,0.07272691,0.1269947,0.6382964],"study_design_scores_gemma":[0.0003904516,0.000522893,0.001397325,0.0001133026,0.0001112516,0.0004742418,0.0005406322,0.9076374,0.01707419,0.04737318,0.02429371,0.00007153687],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.2833592,0.006374145,0.5625475,0.003675075,0.003480024,0.0006933518,0.005679307,0.04604517,0.08814616],"genre_scores_gemma":[0.742129,0.0005844014,0.2274422,0.0004440365,0.0003850898,0.0001480103,0.0101184,0.005155003,0.01359384],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.01940781,"threshold_uncertainty_score":0.06510776,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04670331372071215,"score_gpt":0.3104295284407657,"score_spread":0.2637262147200536,"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."}}