{"id":"W4410427856","doi":"10.1109/tcsvt.2025.3570707","title":"ShiftQuant: Toward Accurate and Efficient Sub-8-bit Integer Training","year":2025,"lang":"en","type":"article","venue":"IEEE Transactions on Circuits and Systems for Video Technology","topic":"Parallel Computing and Optimization Techniques","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"Simon Fraser University","funders":"National Natural Science Foundation of China","keywords":"Computer science; Integer (computer science); Bit (key); Algorithm; Theoretical computer science; Computer network; Programming language","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":[],"consensus_categories":[],"category_scores_codex":[0.0003743223,0.0001914561,0.0003280403,0.00073814,0.000357215,0.0001731406,0.0003605865,0.0002135682,5.53665e-7],"category_scores_gemma":[0.00002315911,0.0001779478,0.00005928801,0.0006230851,0.0001179677,0.0001059669,0.000008683802,0.0002346698,0.000001696319],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00004355199,"about_ca_system_score_gemma":0.00006662393,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0000174688,"about_ca_topic_score_gemma":0.000005028199,"domain_scores_codex":[0.9986555,0.0000464237,0.0003620573,0.0005267771,0.0001019912,0.0003072778],"domain_scores_gemma":[0.9991706,0.000204434,0.00009852321,0.0003534486,0.0001150787,0.00005788891],"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":[0.0000337252,0.0002563553,0.00005051879,0.0005501451,0.0002575894,0.00001671209,0.00277401,0.07516584,0.004005332,0.3809855,0.0007273247,0.5351769],"study_design_scores_gemma":[0.0008969667,0.00034895,0.00002990276,0.0003926149,0.0000372438,0.0001046422,0.0004270591,0.9730543,0.01557532,0.004671091,0.004067788,0.0003940844],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01386703,0.0008454475,0.9817718,0.001134481,0.000689892,0.0005307164,0.00001039762,0.0009262387,0.0002239557],"genre_scores_gemma":[0.9971418,0.0001228633,0.002258623,0.0001456945,0.00001110622,0.0001911099,5.28515e-7,0.00001106231,0.000117191],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9832748,"threshold_uncertainty_score":0.72565,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03626091097901787,"score_gpt":0.2787071509694878,"score_spread":0.24244623999047,"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."}}