StyleRemix: Interpretable Authorship Obfuscation via Distillation and Perturbation of Style Elements
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866929477954568192 |
|---|---|
| author | Fisher, Jillian Hallinan, Skyler Lu, Ximing Gordon, Mitchell Harchaoui, Zaid Choi, Yejin |
| author_facet | Fisher, Jillian Hallinan, Skyler Lu, Ximing Gordon, Mitchell Harchaoui, Zaid Choi, Yejin |
| contents | Authorship obfuscation, rewriting a text to intentionally obscure the identity of the author, is an important but challenging task. Current methods using large language models (LLMs) lack interpretability and controllability, often ignoring author-specific stylistic features, resulting in less robust performance overall.
To address this, we develop StyleRemix, an adaptive and interpretable obfuscation method that perturbs specific, fine-grained style elements of the original input text. StyleRemix uses pre-trained Low Rank Adaptation (LoRA) modules to rewrite an input specifically along various stylistic axes (e.g., formality and length) while maintaining low computational cost. StyleRemix outperforms state-of-the-art baselines and much larger LLMs in a variety of domains as assessed by both automatic and human evaluation.
Additionally, we release AuthorMix, a large set of 30K high-quality, long-form texts from a diverse set of 14 authors and 4 domains, and DiSC, a parallel corpus of 1,500 texts spanning seven style axes in 16 unique directions |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2408_15666 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | StyleRemix: Interpretable Authorship Obfuscation via Distillation and Perturbation of Style Elements Fisher, Jillian Hallinan, Skyler Lu, Ximing Gordon, Mitchell Harchaoui, Zaid Choi, Yejin Computation and Language Authorship obfuscation, rewriting a text to intentionally obscure the identity of the author, is an important but challenging task. Current methods using large language models (LLMs) lack interpretability and controllability, often ignoring author-specific stylistic features, resulting in less robust performance overall. To address this, we develop StyleRemix, an adaptive and interpretable obfuscation method that perturbs specific, fine-grained style elements of the original input text. StyleRemix uses pre-trained Low Rank Adaptation (LoRA) modules to rewrite an input specifically along various stylistic axes (e.g., formality and length) while maintaining low computational cost. StyleRemix outperforms state-of-the-art baselines and much larger LLMs in a variety of domains as assessed by both automatic and human evaluation. Additionally, we release AuthorMix, a large set of 30K high-quality, long-form texts from a diverse set of 14 authors and 4 domains, and DiSC, a parallel corpus of 1,500 texts spanning seven style axes in 16 unique directions |
| title | StyleRemix: Interpretable Authorship Obfuscation via Distillation and Perturbation of Style Elements |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2408.15666 |