{"url":"/method/pattern-exploiting-training","slug":"pattern-exploiting-training","name":"Pattern-Exploiting Training","full_name":"Pattern-Exploiting Training","full_name_withheld":false,"description_markdown":"**Pattern-Exploiting Training** is a semi-supervised training procedure that reformulates input examples as cloze-style phrases to help language models understand a given task. These phrases are then used to assign soft labels to a large set of unlabeled examples. Finally, standard supervised training is performed on the resulting training set. \r\n\r\nIn the case of PET for sentiment classification, first a number of patterns encoding some form of task description are created to convert training examples to cloze questions; for each pattern, a pretrained language model is finetuned. Secondly, the ensemble of trained models annotates unlabeled data. Lastly, a classifier is trained on the resulting soft-labeled dataset.","description_state":"present","introduced_year":null,"introduced_by":{"title":"Exploiting Cloze Questions for Few Shot Text Classification and Natural Language Inference","paper":"/paper/exploiting-cloze-questions-for-few-shot-text","first_author":"Timo Schick","n_authors":2,"url_abs":null,"archive_paper_url":"https://paperswithcode.com/paper/exploiting-cloze-questions-for-few-shot-text"},"source":{"url":"https://arxiv.org/abs/2001.07676v3","title":"Exploiting Cloze Questions for Few Shot Text Classification and Natural Language Inference","url_on_a_paper_host":true},"code_snippet_url":null,"code_snippet_url_on_a_code_host":false,"categories":[{"area":"General","area_id":"general","collection":"Semi-Supervised Learning Methods","url":"/methods/category/semi-supervised-learning-methods","pwc_aliases":[]}],"n_papers_tagged":6,"archive_num_papers":6,"papers_newest_first":[{"paper":null,"title":"Exploring Data Augmentation Methods on Social Media Corpora","date":"2023-03-03","arxiv_id":"2303.02198","n_code_links":0,"syntology":null},{"paper":null,"title":"Enhancing Tabular Reasoning with Pattern Exploiting Training","date":"2022-10-21","arxiv_id":"2210.12259","n_code_links":0,"syntology":null},{"paper":"/paper/quantifying-the-effect-of-color-processing-on","title":"Quantifying the effect of color processing on blood and damaged tissue detection in Whole Slide Images","date":"2022-09-26","arxiv_id":null,"n_code_links":1,"syntology":null},{"paper":null,"title":"Few-shot Named Entity Recognition with Cloze Questions","date":"2021-11-24","arxiv_id":"2111.12421","n_code_links":0,"syntology":null},{"paper":"/paper/exploiting-cloze-questions-for-few-shot-text-1","title":"Exploiting Cloze-Questions for Few-Shot Text Classification and Natural Language Inference","date":"2021-04-01","arxiv_id":null,"n_code_links":1,"syntology":null},{"paper":"/paper/exploiting-cloze-questions-for-few-shot-text","title":"Exploiting Cloze Questions for Few Shot Text Classification and Natural Language Inference","date":"2020-01-21","arxiv_id":"2001.07676","n_code_links":6,"syntology":null}],"papers_shown":6,"tasks":[{"task":"/task/text-classification","name":"Text Classification","papers":3},{"task":"/task/few-shot-learning","name":"Few-Shot Learning","papers":2},{"task":"/task/few-shot-text-classification","name":"Few-Shot Text Classification","papers":2},{"task":"/task/language-modeling","name":"Language Modeling","papers":2},{"task":"/task/language-modelling","name":"Language Modelling","papers":2},{"task":"/task/natural-language-inference","name":"Natural Language Inference","papers":2},{"task":"/task/text-classification-1","name":"text-classification","papers":2},{"task":"/task/artifact-detection","name":"Artifact Detection","papers":1},{"task":"/task/binary-classification","name":"Binary Classification","papers":1},{"task":"/task/blood-detection","name":"Blood Detection","papers":1},{"task":"/task/color-normalization","name":"Color Normalization","papers":1},{"task":"/task/damaged-tissue-detection","name":"Damaged Tissue Detection","papers":1},{"task":"/task/data-augmentation","name":"Data Augmentation","papers":1},{"task":"/task/diagnostic","name":"Diagnostic","papers":1},{"task":"/task/classification","name":"General Classification","papers":1},{"task":"/task/cg","name":"NER","papers":1},{"task":"/task/named-entity-recognition-1","name":"Named Entity Recognition","papers":1},{"task":"/task/named-entity-recognition-ner","name":"Named Entity Recognition (NER)","papers":1},{"task":"/task/transfer-learning","name":"Transfer Learning","papers":1},{"task":"/task/named-entity-recognition","name":"named-entity-recognition","papers":1}],"tasks_shown":20,"n_tasks":21,"usage_by_year":[{"year":"2020","papers":1},{"year":"2021","papers":2},{"year":"2022","papers":2},{"year":"2023","papers":1}],"row_source":"methods_table","archive":{"source":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","archive_url":"https://paperswithcode.com/method/pattern-exploiting-training"},"syntology_read_at":"2026-09-24T18:15:14+00:00"}