{"id":"W4415221973","doi":"10.1109/ms.2025.3621128","title":"Using LLMs to Bridge the Gaps in QA Test Plans at Firefox","year":2025,"lang":"en","type":"article","venue":"IEEE Software","topic":"SAS software applications and methods","field":"Engineering","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Ministère des Ressources naturelles et des Forêts (Québec); Government of Canada; University of Calgary","funders":"","keywords":"Bridge (graph theory); Test (biology); Quality assurance; Reliability (semiconductor); Test plan; Software quality; Process (computing)","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":[0.0001738896,0.0001353025,0.0001432588,0.0001058638,0.0001617764,0.00002123688,0.0002315603,0.0000770367,0.0000245018],"category_scores_gemma":[0.0001712016,0.0001142153,0.00004433819,0.0005838622,0.0000229744,0.00003966777,0.00007228902,0.0001465816,0.00005159113],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002091228,"about_ca_system_score_gemma":0.00003156404,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000151026,"about_ca_topic_score_gemma":0.0003538888,"domain_scores_codex":[0.9992552,0.00002015266,0.0001924456,0.0001858689,0.00008824572,0.0002581086],"domain_scores_gemma":[0.9989967,0.0004909261,0.00001441407,0.0004192289,0.00002660314,0.00005215747],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"not_applicable","study_design_scores_codex":[0.00002951387,0.0001607469,0.5613367,0.0004112608,0.0001146122,0.00003707792,0.004383389,0.2032311,0.03036189,0.0003288635,0.06574333,0.1338615],"study_design_scores_gemma":[0.000962581,0.00004687914,0.4324096,0.0005093926,0.00008275724,0.00003899362,0.000154768,0.009626037,0.04693133,0.00158261,0.506682,0.0009729938],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.7868392,0.0004001186,0.2105483,0.0001667049,0.0005543125,0.0003612178,0.00009458081,0.0004885271,0.0005470361],"genre_scores_gemma":[0.9615716,0.00003914395,0.03670283,0.0006562418,0.0001190555,0.0001628407,0.00001144841,0.00004701778,0.0006897951],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.4409387,"threshold_uncertainty_score":0.4657563,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0331678809201386,"score_gpt":0.3164767833720971,"score_spread":0.2833089024519584,"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."}}