{"id":"W7133713936","doi":"10.1109/ccnis69465.2025.00025","title":"A Practical Recipe for Structured Pruning of MotionGPT: Dependency-Graph Pruning and FFN Channel Reduction","year":2025,"lang":"","type":"article","venue":"","topic":"Graph Theory and Algorithms","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"","keywords":"Recipe; Pruning; Channel (broadcasting); Reduction (mathematics); Pattern recognition (psychology)","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.000607576,0.0009918748,0.0006054825,0.0007467954,0.0006280725,0.0006395662,0.001329245,0.0009445524,0.005047983],"category_scores_gemma":[0.003390727,0.0003877817,0.0008046098,0.0006354719,0.0006738481,0.001147209,0.001484463,0.001612548,0.001561082],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0006269944,"about_ca_system_score_gemma":0.001467841,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004710051,"about_ca_topic_score_gemma":0.01517211,"domain_scores_codex":[0.9996048,0.00009383836,0.00002429683,0.00009513764,0.0001240784,0.00005791704],"domain_scores_gemma":[0.9990446,0.000439221,0.00003749054,0.0002624279,0.0001743767,0.00004190217],"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":[0.0002106458,0.0001645865,0.001922036,0.0002292943,0.00009200654,0.0006722607,0.0004111462,0.2560435,0.04245482,0.05648074,0.02473208,0.6165869],"study_design_scores_gemma":[0.00002876208,0.00004442679,0.0005125281,0.00002348583,0.00003261835,0.0002002294,0.00007364734,0.9292707,0.01729145,0.04380859,0.008693645,0.00001998241],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.01037866,0.00007272231,0.9854599,0.0001406774,0.00003609297,0.00008403254,0.0001695684,0.001497603,0.002160855],"genre_scores_gemma":[0.1749367,0.0001399945,0.817734,0.0002633449,0.00004621426,0.0003385626,0.001131612,0.0008973082,0.004512292],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.005047983,"threshold_uncertainty_score":0.01688719,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02464513646086404,"score_gpt":0.3008554741785991,"score_spread":0.2762103377177351,"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."}}