{"id":"W4385767844","doi":"10.24963/ijcai.2023/561","title":"KEST: Kernel Distance Based Efficient Self-Training for Improving Controllable Text Generation","year":2023,"lang":"en","type":"article","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of British Columbia","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Computer science; Text generation; Generator (circuit theory); Bottleneck; Natural language generation; Fluency; Kernel (algebra); Artificial intelligence; Exploit; Language model; Process (computing); Machine learning; Natural language; Power (physics); Mathematics","routes":{"ca_aff":true,"ca_fund":true,"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":[0.001554326,0.001066984,0.0008500152,0.0007668742,0.0004358225,0.0006587137,0.001806379,0.001144027,0.002898148],"category_scores_gemma":[0.006381124,0.0004536675,0.0007456285,0.0006138798,0.0008101769,0.002210516,0.001917104,0.00176453,0.001693082],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0005901761,"about_ca_system_score_gemma":0.0008368128,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001924163,"about_ca_topic_score_gemma":0.003393757,"domain_scores_codex":[0.9991966,0.000283679,0.00005411889,0.0002187567,0.0001727865,0.00007410243],"domain_scores_gemma":[0.9972296,0.001646115,0.0001886718,0.0004329157,0.0003867593,0.0001159842],"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":[0.0003812411,0.0003479762,0.001989507,0.0002396219,0.00009354814,0.000183093,0.0004014359,0.4104953,0.02534285,0.009289918,0.007704849,0.5435306],"study_design_scores_gemma":[0.00001338484,0.00004665438,0.000107821,0.000005219019,0.000005359303,0.00002774695,0.00001305503,0.9928941,0.003795095,0.002469419,0.0006152734,0.000006904648],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.0293531,0.0003194609,0.9633653,0.0001151392,0.00006484662,0.0000681528,0.00009839278,0.005494122,0.001121582],"genre_scores_gemma":[0.5999564,0.0002508355,0.3885623,0.0004038967,0.0000909286,0.0003254813,0.001522815,0.001461614,0.007425777],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.002898148,"threshold_uncertainty_score":0.009695292,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03806845852147484,"score_gpt":0.2541483938987131,"score_spread":0.2160799353772383,"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."}}