{"id":"W3162734203","doi":"10.18653/v1/2021.findings-acl.84","title":"A Survey of Data Augmentation Approaches for NLP","year":2021,"lang":"en","type":"preprint","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":129,"is_retracted":false,"has_abstract":true,"ca_institutions":"Mila - Quebec Artificial Intelligence Institute","funders":"","keywords":"Computer science; Popularity; Artificial intelligence; Data science; Natural language processing; Resource (disambiguation); Labeled data; Machine learning; Information retrieval","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":[0.0009705182,0.00009052717,0.000187514,0.00004114549,0.00001727956,0.0001263264,0.001948591,0.00008422605,0.000007551249],"category_scores_gemma":[0.0001016459,0.00008739305,0.00003344915,0.00007537546,0.000009650926,0.0002570055,0.003989006,0.00008952143,5.921002e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00001742147,"about_ca_system_score_gemma":0.0002422752,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001339012,"about_ca_topic_score_gemma":0.0006062653,"domain_scores_codex":[0.9987553,0.00008399705,0.0002606429,0.0006368296,0.0001663061,0.00009696886],"domain_scores_gemma":[0.9975756,0.000145808,0.0001350643,0.002005951,0.0001139912,0.00002359529],"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":[0.00001215433,0.000268127,0.004219675,0.001381067,0.000258783,0.000002120369,0.002420848,0.02940329,0.00009776615,0.03625168,0.002229754,0.9234548],"study_design_scores_gemma":[0.00009700818,0.000005492866,0.003731873,0.00002507136,0.000006098345,3.455783e-7,0.00003158953,0.9947132,0.0002852542,0.0009843465,0.00002740838,0.00009230198],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.004604254,0.0001620496,0.9939106,0.0001733086,0.0003411743,0.000290616,0.00006433654,0.00003922788,0.0004144872],"genre_scores_gemma":[0.3569694,0.000006963861,0.6418767,0.00003015396,0.00002524058,0.00002237533,0.0009204573,0.000004543834,0.0001441577],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.9653099,"threshold_uncertainty_score":0.4972011,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.5270259338359118,"score_gpt":0.3715261320235383,"score_spread":0.1554998018123735,"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."}}