{"id":"W4414348545","doi":"10.1109/tse.2025.3611329","title":"DiffGAN: A Test Generation Approach for Differential Testing of Deep Neural Networks for Image Analysis","year":2025,"lang":"en","type":"article","venue":"IEEE Transactions on Software Engineering","topic":"Image Processing and 3D Reconstruction","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":false,"ca_institutions":"École de Technologie Supérieure; University of Ottawa","funders":"Natural Sciences and Engineering Research Council of Canada; Science Foundation Ireland; General Motors Corporation","keywords":"Artificial neural network; Image (mathematics); Pattern recognition (psychology); Test (biology); Image processing; Differential (mechanical device)","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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.00009124303,0.0001521667,0.0002276478,0.0003823364,0.0002059772,0.0001181296,0.0001992836,0.000076843,9.877451e-7],"category_scores_gemma":[0.00007895888,0.0001561041,0.0002018793,0.0009459483,0.00002000056,0.0002483491,0.000002208275,0.0001235122,6.918125e-8],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00003620666,"about_ca_system_score_gemma":0.00002219186,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000004111302,"about_ca_topic_score_gemma":0.000002346797,"domain_scores_codex":[0.999114,0.000009598511,0.0002695638,0.0003105018,0.0000864364,0.0002098912],"domain_scores_gemma":[0.9991144,0.0003950145,0.00007134076,0.0002134947,0.0001711072,0.00003466699],"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":[0.000005365441,0.00006400288,0.00005922867,0.0001319917,0.0001129998,1.320061e-7,0.00004039293,0.8732749,0.003970628,0.00002416597,0.000004227023,0.122312],"study_design_scores_gemma":[0.0003265614,0.00004835709,0.0001039793,0.00001624548,0.0002250669,0.000003631728,0.000005271381,0.9776283,0.02150109,0.000008951327,0.000002189763,0.0001303681],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.003546826,0.00004049586,0.9951823,0.00001143485,0.0006698469,0.000253311,0.00001444377,0.0002772527,0.000004056835],"genre_scores_gemma":[0.5359986,7.020723e-7,0.463803,0.000006427059,0.00004850255,0.0001070072,0.000006706163,0.00000799975,0.00002113139],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.5324517,"threshold_uncertainty_score":0.6365741,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01220046323444045,"score_gpt":0.2224542094514625,"score_spread":0.210253746217022,"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."}}