{"url":"/task/text-style-transfoer","name":"Text Style Transfer","slug":"text-style-transfoer","description_markdown":"Text Style Transfer is the task of controlling certain attributes of generated text. The state-of-the-art methods can be categorized into two main types which are used on parallel and non-parallel data. Methods on parallel data are typically supervised methods that use a neural sequence-to-sequence model with the encoder-decoder architecture. Methods on non-parallel data are usually unsupervised approaches using Disentanglement, Prototype Editing and Pseudo-Parallel Corpus Construction.\r\n\r\nThe popular benchmark for this task is  the Yelp Review Dataset. Models are typically evaluated with the metrics of Sentiment Accuracy, BLEU, and PPL.","categories":[{"name":"Natural Language Processing","url":"/area/natural-language-processing"}],"source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","slug_source":"archive_url"},"counts":{"papers_tagged":186,"papers_with_code":93,"benchmarks":2,"benchmark_tables_in_archive":2,"benchmark_tables_shown":2,"benchmark_tables_withheld_as_spam":0,"benchmark_definition":"a leaderboard table with at least one row; benchmark_tables_shown also counts the zero-row tables; benchmark_tables_in_archive adds the tables withheld as spam","datasets":7,"subtasks":3,"parent_tasks":2},"benchmarks":[{"leaderboard":"/sota/text-style-transfer-on-yelp-review-dataset","slug":"text-style-transfer-on-yelp-review-dataset","dataset":"Yelp Review Dataset (Small)","dataset_url":"/dataset/yelp","rows_in_archive":8,"metrics":["G-Score (BLEU, Accuracy)"],"first_row_in_archive_order":{"model":"SAE+Discriminator","paper_title":"Style Transfer for Texts: Retrain, Report Errors, Compare with Rewrites","paper_url":"/paper/style-transfer-for-texts-to-err-is-human-but","paper_date":"2019-08-19","arxiv_id":"1908.06809","code_links":[{"title":"VAShibaev/text_style_transfer","url":"https://github.com/VAShibaev/text_style_transfer"}],"syntology":{"n":1,"n_ran":0,"n_unverified":1,"n_pointer_only":0}}},{"leaderboard":"/sota/text-style-transfer-on-yelp-review-dataset-1","slug":"text-style-transfer-on-yelp-review-dataset-1","dataset":"Yelp Review Dataset (Large)","dataset_url":"/dataset/yelp","rows_in_archive":2,"metrics":["G-Score (BLEU, Accuracy)","BLEU"],"first_row_in_archive_order":{"model":"SentiInc","paper_title":"SentiInc: Incorporating Sentiment Information into Sentiment Transfer Without Parallel Data","paper_url":"/paper/sentiinc-incorporating-sentiment-information","paper_date":"2020-04-08","arxiv_id":null,"code_links":[],"syntology":null}}],"datasets":[{"url":"/dataset/yelp","name":"Yelp","full_name":"","num_papers_in_archive":86},{"url":"/dataset/touchdown-dataset","name":"Touchdown Dataset","full_name":null,"num_papers_in_archive":19},{"url":"/dataset/textseg","name":"TextSeg","full_name":null,"num_papers_in_archive":16},{"url":"/dataset/styleptb","name":"StylePTB","full_name":"","num_papers_in_archive":11},{"url":"/dataset/xformal","name":"XFORMAL","full_name":"","num_papers_in_archive":10},{"url":"/dataset/stylegallery","name":"StyleGallery","full_name":"","num_papers_in_archive":7},{"url":"/dataset/textbox-2-0","name":"TextBox 2.0","full_name":"","num_papers_in_archive":3}],"subtasks":[{"url":"/task/formality-style-transfer","name":"Formality Style Transfer"},{"url":"/task/semi-supervised-formality-style-transfer","name":"Semi-Supervised Formality Style Transfer"},{"url":"/task/word-attribute-transfer","name":"Word Attribute Transfer"}],"parent_tasks":[{"url":"/task/2d-semantic-segmentation","name":"2D Semantic Segmentation"},{"url":"/task/text-generation","name":"Text Generation"}],"papers":{"order":"repositories listed in the archive (desc), then date (desc); the archive holds no stars","population":"papers tagged with this task that list at least one repository in the archive","shown":30,"of":93,"tagged_in_all":186,"items":[{"url":"/paper/style-transfer-from-non-parallel-text-by","title":"Style Transfer from Non-Parallel Text by Cross-Alignment","date":"2017-05-26","arxiv_id":"1705.09655","repositories_listed":12,"syntology":{"n":7,"n_ran":6,"n_unverified":1,"n_pointer_only":6}},{"url":"/paper/a-probabilistic-formulation-of-unsupervised-1","title":"A Probabilistic Formulation of Unsupervised Text Style Transfer","date":"2020-02-10","arxiv_id":"2002.03912","repositories_listed":5,"syntology":{"n":25,"n_ran":18,"n_unverified":7,"n_pointer_only":12}},{"url":"/paper/vae-based-text-style-transfer-with-pivot","title":"VAE based Text Style Transfer with Pivot Words Enhancement Learning","date":"2021-12-06","arxiv_id":"2112.03154","repositories_listed":4,"syntology":null},{"url":"/paper/190505621","title":"Style Transformer: Unpaired Text Style Transfer without Disentangled Latent Representation","date":"2019-05-14","arxiv_id":"1905.05621","repositories_listed":4,"syntology":{"n":3,"n_ran":3,"n_unverified":0,"n_pointer_only":3}},{"url":"/paper/it5-large-scale-text-to-text-pretraining-for","title":"IT5: Text-to-text Pretraining for Italian Language Understanding and Generation","date":"2022-03-07","arxiv_id":"2203.03759","repositories_listed":3,"syntology":null},{"url":"/paper/beyond-fully-connected-layers-with","title":"Beyond Fully-Connected Layers with Quaternions: Parameterization of Hypercomplex Multiplications with $1/n$ Parameters","date":"2021-02-17","arxiv_id":"2102.08597","repositories_listed":3,"syntology":{"n":1,"n_ran":0,"n_unverified":1,"n_pointer_only":0}},{"url":"/paper/latent-space-secrets-of-denoising-text","title":"Educating Text Autoencoders: Latent Representation Guidance via Denoising","date":"2019-05-29","arxiv_id":"1905.12777","repositories_listed":3,"syntology":null},{"url":"/paper/unsupervised-text-style-transfer-via","title":"IMaT: Unsupervised Text Attribute Transfer via Iterative Matching and Translation","date":"2019-01-31","arxiv_id":"1901.11333","repositories_listed":3,"syntology":null},{"url":"/paper/multiple-attribute-text-style-transfer","title":"Multiple-Attribute Text Style Transfer","date":"2018-11-01","arxiv_id":"1811.00552","repositories_listed":3,"syntology":null},{"url":"/paper/disentangled-representation-learning-for-non","title":"Disentangled Representation Learning for Non-Parallel Text Style Transfer","date":"2018-08-13","arxiv_id":"1808.04339","repositories_listed":3,"syntology":{"n":3,"n_ran":3,"n_unverified":0,"n_pointer_only":3}},{"url":"/paper/style-transfer-through-back-translation","title":"Style Transfer Through Back-Translation","date":"2018-04-24","arxiv_id":"1804.09000","repositories_listed":3,"syntology":null},{"url":"/paper/multilingual-text-style-transfer-datasets","title":"Multilingual Text Style Transfer: Datasets & Models for Indian Languages","date":"2024-05-31","arxiv_id":"2405.20805","repositories_listed":2,"syntology":null},{"url":"/paper/tsst-a-benchmark-and-evaluation-models-for","title":"PSST: A Benchmark for Evaluation-driven Text Public-Speaking Style Transfer","date":"2023-11-14","arxiv_id":"2311.08389","repositories_listed":2,"syntology":null},{"url":"/paper/studying-the-role-of-named-entities-for","title":"Studying the role of named entities for content preservation in text style transfer","date":"2022-06-20","arxiv_id":"2206.09676","repositories_listed":2,"syntology":null},{"url":"/paper/styleptb-a-compositional-benchmark-for-fine","title":"StylePTB: A Compositional Benchmark for Fine-grained Controllable Text Style Transfer","date":"2021-04-12","arxiv_id":"2104.05196","repositories_listed":2,"syntology":null},{"url":"/paper/deep-learning-for-text-attribute-transfer-a","title":"Deep Learning for Text Style Transfer: A Survey","date":"2020-11-01","arxiv_id":"2011.00416","repositories_listed":2,"syntology":null},{"url":"/paper/text-style-transfer-a-review-and-experiment","title":"Text Style Transfer: A Review and Experimental Evaluation","date":"2020-10-24","arxiv_id":"2010.12742","repositories_listed":2,"syntology":null},{"url":"/paper/a-dual-reinforcement-learning-framework-for","title":"A Dual Reinforcement Learning Framework for Unsupervised Text Style Transfer","date":"2019-05-24","arxiv_id":"1905.10060","repositories_listed":2,"syntology":{"n":4,"n_ran":0,"n_unverified":4,"n_pointer_only":0}},{"url":"/paper/structured-content-preservation-for","title":"Structured Content Preservation for Unsupervised Text Style Transfer","date":"2018-10-15","arxiv_id":"1810.06526","repositories_listed":2,"syntology":{"n":4,"n_ran":0,"n_unverified":4,"n_pointer_only":0}},{"url":"/paper/style-transfer-in-text-exploration-and","title":"Style Transfer in Text: Exploration and Evaluation","date":"2017-11-18","arxiv_id":"1711.06861","repositories_listed":2,"syntology":null},{"url":"/paper/multilingual-and-explainable-text","title":"Multilingual and Explainable Text Detoxification with Parallel Corpora","date":"2024-12-16","arxiv_id":"2412.11691","repositories_listed":1,"syntology":null},{"url":"/paper/style-specific-neurons-for-steering-llms-in","title":"Style-Specific Neurons for Steering LLMs in Text Style Transfer","date":"2024-10-01","arxiv_id":"2410.00593","repositories_listed":1,"syntology":{"n":3,"n_ran":2,"n_unverified":1,"n_pointer_only":3}},{"url":"/paper/was-dataset-and-methods-for-artistic-text","title":"WAS: Dataset and Methods for Artistic Text Segmentation","date":"2024-07-31","arxiv_id":"2408.00106","repositories_listed":1,"syntology":null},{"url":"/paper/hard-prompts-made-interpretable-sparse","title":"Hard Prompts Made Interpretable: Sparse Entropy Regularization for Prompt Tuning with RL","date":"2024-07-20","arxiv_id":"2407.14733","repositories_listed":1,"syntology":{"n":2,"n_ran":1,"n_unverified":1,"n_pointer_only":0}},{"url":"/paper/tinystyler-efficient-few-shot-text-style","title":"TinyStyler: Efficient Few-Shot Text Style Transfer with Authorship Embeddings","date":"2024-06-21","arxiv_id":"2406.15586","repositories_listed":1,"syntology":null},{"url":"/paper/style-transfer-with-multi-iteration","title":"Style Transfer with Multi-iteration Preference Optimization","date":"2024-06-17","arxiv_id":"2406.11581","repositories_listed":1,"syntology":{"n":2,"n_ran":2,"n_unverified":0,"n_pointer_only":0}},{"url":"/paper/are-large-language-models-actually-good-at","title":"Are Large Language Models Actually Good at Text Style Transfer?","date":"2024-06-09","arxiv_id":"2406.05885","repositories_listed":1,"syntology":null},{"url":"/paper/text-detoxification-as-style-transfer-in","title":"Text Detoxification as Style Transfer in English and Hindi","date":"2024-02-12","arxiv_id":"2402.07767","repositories_listed":1,"syntology":null},{"url":"/paper/cat-llm-prompting-large-language-models-with","title":"CAT-LLM: Prompting Large Language Models with Text Style Definition for Chinese Article-style Transfer","date":"2024-01-11","arxiv_id":"2401.05707","repositories_listed":1,"syntology":null},{"url":"/paper/steer-unified-style-transfer-with-expert","title":"STEER: Unified Style Transfer with Expert Reinforcement","date":"2023-11-13","arxiv_id":"2311.07167","repositories_listed":1,"syntology":null}],"syntology_records":10,"syntology_note":"a paper without a record is not a recorded non-run: it may lack an arXiv id or simply be absent from the graph layer"},"description_links":{"kept":0,"unwrapped_to_text":0,"bare_urls_linked":0,"relative_images_dropped":0,"rule":"internal links are kept only when the target slug exists in the catalog"},"syntology":{"read_at":"2026-09-25T09:33:49+00:00","claim":"Per-sample execution status on synthesized fixtures ('ran N of M samples'); not a correctness claim and not a ranking signal.","status_vocabulary":{"ran_honours":"ran, honoured the contract we drafted","ran_violates":"ran, violated the contract we drafted","ran_draft_wrong":"ran; our contract draft was wrong, not the code","ran_fixture":"ran; our fixture could not drive it","ran":"ran on a synthesized input","unverified":"unverified (harvested, no recorded run)"}},"not_shown":{"libraries":"the archive has no per-task library table","trend_sparklines":"the Trend column of the benchmarks table was a rendered image; it is not in the archive","social_and_latest_sorts":"stars and social signals are not in the archive"}}