{"id":"W4210768747","doi":"10.1145/3472393","title":"RD-IOD: Two-Level Residual-Distillation-Based Triple-Network for Incremental Object Detection","year":2022,"lang":"en","type":"article","venue":"ACM Transactions on Multimedia Computing Communications and Applications","topic":"Advanced Neural Network Applications","field":"Computer Science","cited_by":25,"is_retracted":false,"has_abstract":true,"ca_institutions":"Carleton University","funders":"Beijing Municipal Science and Technology Commission; National Natural Science Foundation of China","keywords":"Computer science; Residual; Artificial intelligence; Object (grammar); Convolutional neural network; Forgetting; Distillation; Machine learning; Feature (linguistics); Learning object; Object detection; Pattern recognition (psychology); Algorithm","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","sts"],"consensus_categories":[],"category_scores_codex":[0.0006188553,0.0003178209,0.0002844102,0.0002939596,0.006202996,0.00014988,0.002522474,0.00007168432,0.00001785023],"category_scores_gemma":[0.00005174212,0.0003818277,0.0001537609,0.001625376,0.0002157162,0.000228818,0.0003069522,0.0006461099,0.00001512474],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002592304,"about_ca_system_score_gemma":0.000137984,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00006945733,"about_ca_topic_score_gemma":0.0001700473,"domain_scores_codex":[0.9973968,0.0002974586,0.0006706805,0.0007974482,0.0003620156,0.0004756566],"domain_scores_gemma":[0.9926399,0.003161219,0.0003723455,0.003438937,0.0002091321,0.0001784857],"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":[0.00004586411,0.0006882583,0.0001088074,0.00001752015,0.00007728667,3.660612e-7,0.0003601494,0.2431931,0.002765062,0.009851765,0.0003151497,0.7425767],"study_design_scores_gemma":[0.001603598,0.0001818455,0.0006123142,0.00001570201,0.00005344371,0.0000177355,0.0001449224,0.9394935,0.001142984,0.006624824,0.04963516,0.0004739442],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.0006263848,0.0002745831,0.9906759,0.004642393,0.0001828651,0.002569489,0.0002930074,0.0005967729,0.000138584],"genre_scores_gemma":[0.5326507,0.00005345247,0.4628532,0.0004069035,0.000101922,0.003652021,0.0001985378,0.00003112849,0.00005212296],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.7421027,"threshold_uncertainty_score":0.9998634,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04402762313761858,"score_gpt":0.3098828924033286,"score_spread":0.26585526926571,"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."}}