{"id":"W4416341683","doi":"10.1109/socc66126.2025.11235380","title":"SparHiXcel: A Cost-Effective Sparsity-Aware Convolutional Accelerator","year":2025,"lang":"","type":"article","venue":"","topic":"Advanced Neural Network Applications","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Institut National de la Recherche Scientifique","funders":"","keywords":"Field-programmable gate array; Convolutional neural network; Reduction (mathematics); Hardware acceleration; Exploit; Efficient energy use; Energy (signal processing); Deep learning; Filter (signal processing)","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","insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.000258278,0.0005388066,0.0005297165,0.000290451,0.001016397,0.0003935718,0.001895187,0.000253022,0.0004223303],"category_scores_gemma":[0.0000712816,0.0005730076,0.000238428,0.002932023,0.000413309,0.0009624727,0.001565196,0.0006504494,0.000912605],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0005302531,"about_ca_system_score_gemma":0.0005874207,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00003272912,"about_ca_topic_score_gemma":0.00006212195,"domain_scores_codex":[0.9961355,0.0002327023,0.0006233653,0.001578597,0.0004746846,0.0009551188],"domain_scores_gemma":[0.9965531,0.0008481992,0.0002393273,0.001427396,0.0005955126,0.000336519],"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":[0.00006615052,0.0004853122,0.002775664,0.00006006591,0.0001729306,0.00001987286,0.000130153,0.003040973,0.0005219874,0.7274601,0.04875866,0.2165081],"study_design_scores_gemma":[0.001923917,0.0001799142,0.020491,0.0002116873,0.0001095975,0.00003490925,0.00006271699,0.5659193,0.007797822,0.03628673,0.3659084,0.001073931],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.001221633,0.0007698527,0.9651849,0.008497936,0.001635159,0.00534698,0.00004784813,0.0004544847,0.0168412],"genre_scores_gemma":[0.9585707,0.0001947655,0.0172212,0.005810011,0.0003542472,0.002100958,0.0000240891,0.00002779214,0.01569629],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.957349,"threshold_uncertainty_score":0.9998653,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02592344794662118,"score_gpt":0.3069476937123321,"score_spread":0.2810242457657109,"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."}}