{"id":"W7131074035","doi":"10.1109/iccvw69036.2025.00422","title":"Enhancing Generalization in Data-Free Quantization via Mixup-Class Prompting","year":2025,"lang":"","type":"article","venue":"","topic":"Advanced Data Compression Techniques","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Hyperion Technologies (Canada)","funders":"Institute for Information and Communications Technology Promotion; Ministry of Science and ICT, South Korea","keywords":"Quantization (signal processing); Generalizability theory; Generative grammar; Synthetic data; Convolutional neural network; Vector quantization; Adversarial system; Generalization; Source code","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","open_science"],"consensus_categories":["open_science"],"category_scores_codex":[0.001399385,0.0004314355,0.0004850694,0.0008594345,0.0003284456,0.000535366,0.005819018,0.0002596279,0.00008838315],"category_scores_gemma":[0.001014958,0.0004554113,0.00004673275,0.003090116,0.00008742746,0.004786501,0.009021967,0.0003995982,0.0000255735],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002467521,"about_ca_system_score_gemma":0.000356819,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0003166845,"about_ca_topic_score_gemma":0.001262732,"domain_scores_codex":[0.9950135,0.0004227853,0.001458309,0.001847166,0.0005878021,0.0006704306],"domain_scores_gemma":[0.9934618,0.0002334277,0.0004829778,0.005426401,0.0002983784,0.00009700502],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00005239921,0.0006639203,0.003365892,0.0007295427,0.00005874657,0.00003835484,0.0007392024,0.008371896,0.1824777,0.2668168,0.02654891,0.5101366],"study_design_scores_gemma":[0.0004907955,0.00002706917,0.000425934,0.0008507316,0.00001340478,0.000003483217,0.0000256822,0.8225119,0.1518512,0.01583162,0.007573854,0.0003943059],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.0006419421,0.0006499282,0.9921806,0.001956419,0.001035958,0.0009592812,0.00003078911,0.0006370337,0.001908006],"genre_scores_gemma":[0.1618218,0.0003856662,0.8339618,0.001224001,0.0001277087,0.00006519625,0.0004240909,0.00003837784,0.001951401],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.81414,"threshold_uncertainty_score":0.9997898,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03167382510845623,"score_gpt":0.3309409491838378,"score_spread":0.2992671240753815,"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."}}