TSTBench: A Comprehensive Benchmark for Text Style Transfer
In recent years, researchers in computational linguistics have shown a growing interest in the style of text, with a specific focus on the text style transfer (TST) task. While numerous innovative methods have been proposed, it has been observed that the existing evaluations are insufficient to vali...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-05-01
|
| Series: | Entropy |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1099-4300/27/6/575 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849432961297940480 |
|---|---|
| author | Yifei Xie Jiaping Gui Zhengping Che Leqian Zhu Yahao Hu Zhisong Pan |
| author_facet | Yifei Xie Jiaping Gui Zhengping Che Leqian Zhu Yahao Hu Zhisong Pan |
| author_sort | Yifei Xie |
| collection | DOAJ |
| description | In recent years, researchers in computational linguistics have shown a growing interest in the style of text, with a specific focus on the text style transfer (TST) task. While numerous innovative methods have been proposed, it has been observed that the existing evaluations are insufficient to validate the claims and precisely measure the performance. This challenge primarily stems from rapid advancements and diverse settings of these methods, with the associated (re)implementation and reproducibility hurdles. To bridge this gap, we introduce a comprehensive benchmark for TST known as <b>TSTBench</b>. TSTBench includes a codebase encompassing implementations of 13 state-of-the-art algorithms and a standardized protocol for text style transfer. Based on the codebase and protocol, we have conducted thorough experiments across seven datasets, resulting in a total of 7000+ evaluations. Our work provides extensive analysis from various perspectives, explores the performance of representative baselines across various datasets, and offers insights into the task and evaluation processes to guide future research in TST. |
| format | Article |
| id | doaj-art-6f8addc7bb2b4ceab66f7f948e808e80 |
| institution | Kabale University |
| issn | 1099-4300 |
| language | English |
| publishDate | 2025-05-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Entropy |
| spelling | doaj-art-6f8addc7bb2b4ceab66f7f948e808e802025-08-20T03:27:14ZengMDPI AGEntropy1099-43002025-05-0127657510.3390/e27060575TSTBench: A Comprehensive Benchmark for Text Style TransferYifei Xie0Jiaping Gui1Zhengping Che2Leqian Zhu3Yahao Hu4Zhisong Pan5Command and Control Engineering College, Army Engineering University, Nanjing 210007, ChinaSchool of Computer Science, Shanghai Jiaotong University, Shanghai 200240, ChinaBeijing Innovation Center of Humanoid Robotics, Beijing 100176, ChinaCommand and Control Engineering College, Army Engineering University, Nanjing 210007, ChinaCommand and Control Engineering College, Army Engineering University, Nanjing 210007, ChinaCommand and Control Engineering College, Army Engineering University, Nanjing 210007, ChinaIn recent years, researchers in computational linguistics have shown a growing interest in the style of text, with a specific focus on the text style transfer (TST) task. While numerous innovative methods have been proposed, it has been observed that the existing evaluations are insufficient to validate the claims and precisely measure the performance. This challenge primarily stems from rapid advancements and diverse settings of these methods, with the associated (re)implementation and reproducibility hurdles. To bridge this gap, we introduce a comprehensive benchmark for TST known as <b>TSTBench</b>. TSTBench includes a codebase encompassing implementations of 13 state-of-the-art algorithms and a standardized protocol for text style transfer. Based on the codebase and protocol, we have conducted thorough experiments across seven datasets, resulting in a total of 7000+ evaluations. Our work provides extensive analysis from various perspectives, explores the performance of representative baselines across various datasets, and offers insights into the task and evaluation processes to guide future research in TST.https://www.mdpi.com/1099-4300/27/6/575text style transferlarge language models (LLMs)text generationtransformerbenchmarkdeep learning |
| spellingShingle | Yifei Xie Jiaping Gui Zhengping Che Leqian Zhu Yahao Hu Zhisong Pan TSTBench: A Comprehensive Benchmark for Text Style Transfer Entropy text style transfer large language models (LLMs) text generation transformer benchmark deep learning |
| title | TSTBench: A Comprehensive Benchmark for Text Style Transfer |
| title_full | TSTBench: A Comprehensive Benchmark for Text Style Transfer |
| title_fullStr | TSTBench: A Comprehensive Benchmark for Text Style Transfer |
| title_full_unstemmed | TSTBench: A Comprehensive Benchmark for Text Style Transfer |
| title_short | TSTBench: A Comprehensive Benchmark for Text Style Transfer |
| title_sort | tstbench a comprehensive benchmark for text style transfer |
| topic | text style transfer large language models (LLMs) text generation transformer benchmark deep learning |
| url | https://www.mdpi.com/1099-4300/27/6/575 |
| work_keys_str_mv | AT yifeixie tstbenchacomprehensivebenchmarkfortextstyletransfer AT jiapinggui tstbenchacomprehensivebenchmarkfortextstyletransfer AT zhengpingche tstbenchacomprehensivebenchmarkfortextstyletransfer AT leqianzhu tstbenchacomprehensivebenchmarkfortextstyletransfer AT yahaohu tstbenchacomprehensivebenchmarkfortextstyletransfer AT zhisongpan tstbenchacomprehensivebenchmarkfortextstyletransfer |