{"id":"W4388184947","doi":"10.48550/arxiv.2310.20558","title":"Breaking the Token Barrier: Chunking and Convolution for Efficient Long Text Classification with BERT","year":2023,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Topic Modeling","field":"Computer Science","cited_by":4,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Natural Sciences and Engineering Research Council of Canada; Canadian Institutes of Health Research; Alliance de recherche numérique du Canada","keywords":"Computer science; Chunking (psychology); Inference; Security token; Artificial intelligence; Benchmark (surveying); Machine learning","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0009354651,0.0009209899,0.0006287112,0.000626028,0.0004529482,0.0009522707,0.001672331,0.0008758699,0.003425216],"category_scores_gemma":[0.003133574,0.0004472166,0.000607073,0.0006926304,0.0007328801,0.00427313,0.001269638,0.001906535,0.002106048],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001072053,"about_ca_system_score_gemma":0.001093496,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.006970503,"about_ca_topic_score_gemma":0.01222159,"domain_scores_codex":[0.9997314,0.00005872869,0.00001663764,0.00009918831,0.00004580185,0.00004820995],"domain_scores_gemma":[0.9991878,0.0003579634,0.00006641541,0.0002345232,0.00009894226,0.0000542405],"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":[0.0008945917,0.0002371125,0.004162774,0.0002481614,0.0001346621,0.0003266307,0.0004929982,0.327659,0.03383897,0.04233721,0.01957116,0.5700968],"study_design_scores_gemma":[0.00001130092,0.00004217525,0.0002030834,0.00001146111,0.00001360208,0.00004439772,0.00002799662,0.9735857,0.007128211,0.01681894,0.002102454,0.00001067263],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.06728718,0.0009599788,0.9163858,0.0005914382,0.0002381725,0.00007287907,0.000555287,0.01025325,0.003656029],"genre_scores_gemma":[0.7499426,0.0005439045,0.235826,0.0004960825,0.0001494826,0.0001373125,0.002031277,0.0007446554,0.01012877],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.006970503,"threshold_uncertainty_score":0.01385987,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1038989451188537,"score_gpt":0.199709248229643,"score_spread":0.09581030311078935,"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."}}