{"id":"W7092205149","doi":"10.1145/3725843.3756105","title":"TransFusion: End-to-End Transformer Acceleration via Graph Fusion and Pipelining","year":2025,"lang":"","type":"article","venue":"","topic":"Advanced Graph Neural Networks","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta","funders":"","keywords":"Transformer; Graph; Fusion; Acceleration; Minification","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"],"consensus_categories":[],"category_scores_codex":[0.0005072166,0.0006238861,0.0005695462,0.0007485197,0.00104377,0.0005266083,0.0008189945,0.0003399947,0.0004690647],"category_scores_gemma":[0.00002017546,0.0005899317,0.0002443083,0.003416016,0.0002025015,0.001746345,0.0002005373,0.0006780901,0.00002671289],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0000512146,"about_ca_system_score_gemma":0.0001183171,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0000919472,"about_ca_topic_score_gemma":0.0004949104,"domain_scores_codex":[0.995698,0.0001776243,0.001004049,0.001609062,0.0005996326,0.0009116497],"domain_scores_gemma":[0.9982041,0.0002975166,0.0001108348,0.0007577913,0.0002109943,0.0004188016],"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":[0.000164,0.0000931738,0.0002779285,0.00005936588,0.00003524444,0.00001540495,0.001479837,0.002009579,0.0215596,0.01306832,0.0004125304,0.960825],"study_design_scores_gemma":[0.005690615,0.00239459,0.01151821,0.001303847,0.0003168309,0.0001278659,0.0005095758,0.79162,0.1027397,0.05644996,0.02431916,0.00300965],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.03839409,0.001447995,0.9405662,0.008004995,0.00212824,0.001151439,0.000004565196,0.0002322358,0.008070193],"genre_scores_gemma":[0.9560694,0.003602729,0.03435198,0.004057728,0.0001635052,0.00004894302,0.000007356834,0.00002854661,0.001669789],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9578153,"threshold_uncertainty_score":0.9996552,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01449664772490705,"score_gpt":0.2640658717116467,"score_spread":0.2495692239867397,"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."}}