{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/virtual-data-augmentation-a-robust-and","title":"Virtual Data Augmentation: A Robust and General Framework for Fine-tuning Pre-trained Models","arxiv_id":"2109.05793","date":"2021-09-13","proceeding":"EMNLP 2021 11","authors":["Kun Zhou","Wayne Xin Zhao","Sirui Wang","Fuzheng Zhang","Wei Wu","Ji-Rong Wen"],"abstract":"Recent works have shown that powerful pre-trained language models (PLM) can be fooled by small perturbations or intentional attacks. To solve this issue, various data augmentation techniques are proposed to improve the robustness of PLMs. However, it is still challenging to augment semantically relevant examples with sufficient diversity. In this work, we present Virtual Data Augmentation (VDA), a general framework for robustly fine-tuning PLMs. Based on the original token embeddings, we construct a multinomial mixture for augmenting virtual data embeddings, where a masked language model guarantees the semantic relevance and the Gaussian noise provides the augmentation diversity. Furthermore, a regularized training strategy is proposed to balance the two aspects. Extensive experiments on six datasets show that our approach is able to improve the robustness of PLMs and alleviate the performance degradation under adversarial attacks. Our codes and data are publicly available at \\textcolor{blue}{\\url{https://github.com/RUCAIBox/VDA}}.","url_abs":"https://arxiv.org/abs/2109.05793v1","url_pdf":"https://arxiv.org/pdf/2109.05793v1.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"virtual-data-augmentation-a-robust-and","repo_url":"https://github.com/rucaibox/vda","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"data-augmentation","task_name":"Data Augmentation"},{"task_slug":"diversity","task_name":"Diversity"},{"task_slug":"language-modeling","task_name":"Language Modeling"},{"task_slug":"language-modelling","task_name":"Language Modelling"}],"methods":[{"method_slug":"virtual-data-augmentation","method_name":"Virtual Data Augmentation"}],"datasets_introduced":[],"methods_introduced":[{"slug":"virtual-data-augmentation","name":"Virtual Data Augmentation","full_name":"Virtual Data Augmentation"}],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/2109.05793","atlas_url":"https://app.syntology.ai/?focus=2109.05793","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}