{"id":"W4416674895","doi":"10.7717/peerj-cs.3388","title":"Towards optimal sparse CNNs: sparsity-friendly knowledge distillation through feature decoupling","year":2025,"lang":"en","type":"article","venue":"PeerJ Computer Science","topic":"Advanced Neural Network Applications","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Pooling; Feature (linguistics); Decoupling (probability); Distillation; Artificial neural network; Convolutional neural network","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":false,"about_ca":true,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0005108654,0.0002931795,0.0002762525,0.0002635836,0.001031073,0.0006230677,0.003289894,0.00008537537,0.000003767274],"category_scores_gemma":[0.00006798036,0.0002822473,0.00009760325,0.004201121,0.0004752269,0.002083566,0.002083105,0.0003263628,0.00007912141],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002282892,"about_ca_system_score_gemma":0.0004473939,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001074234,"about_ca_topic_score_gemma":0.000009962931,"domain_scores_codex":[0.997045,0.00003910578,0.0003180858,0.001320139,0.0005608423,0.000716817],"domain_scores_gemma":[0.9977099,0.0001714797,0.0001487722,0.001348848,0.0004490164,0.0001719941],"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":[0.00000994787,0.0001886624,0.0007375695,0.00003397725,0.00001810997,0.00001426739,0.001291407,0.1061023,0.001161875,0.5535439,0.01019644,0.3267015],"study_design_scores_gemma":[0.0002474681,0.0000564426,0.008032227,0.00005919073,0.000009244824,0.00002311395,0.000008390939,0.9395483,0.003033279,0.01249233,0.0361503,0.0003397684],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.01133058,0.0003693219,0.977956,0.003717148,0.001659332,0.0003605219,0.000003165926,0.0005276187,0.004076332],"genre_scores_gemma":[0.3826895,0.00002246473,0.6162841,0.0004080498,0.000192931,0.00002495829,0.000003450788,0.000007812111,0.0003667671],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.8334459,"threshold_uncertainty_score":0.999963,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02110623592090117,"score_gpt":0.3079105817514071,"score_spread":0.286804345830506,"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."}}