Camouflage is all you need: Evaluating and Enhancing Language Model Robustness Against Camouflage Adversarial Attacks
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_ | 1866913235071926272 |
|---|---|
| author | Huertas-García, Álvaro Martín, Alejandro Huertas-Tato, Javier Camacho, David |
| author_facet | Huertas-García, Álvaro Martín, Alejandro Huertas-Tato, Javier Camacho, David |
| contents | Adversarial attacks represent a substantial challenge in Natural Language Processing (NLP). This study undertakes a systematic exploration of this challenge in two distinct phases: vulnerability evaluation and resilience enhancement of Transformer-based models under adversarial attacks.
In the evaluation phase, we assess the susceptibility of three Transformer configurations, encoder-decoder, encoder-only, and decoder-only setups, to adversarial attacks of escalating complexity across datasets containing offensive language and misinformation. Encoder-only models manifest a 14% and 21% performance drop in offensive language detection and misinformation detection tasks, respectively. Decoder-only models register a 16% decrease in both tasks, while encoder-decoder models exhibit a maximum performance drop of 14% and 26% in the respective tasks.
The resilience-enhancement phase employs adversarial training, integrating pre-camouflaged and dynamically altered data. This approach effectively reduces the performance drop in encoder-only models to an average of 5% in offensive language detection and 2% in misinformation detection tasks. Decoder-only models, occasionally exceeding original performance, limit the performance drop to 7% and 2% in the respective tasks. Although not surpassing the original performance, Encoder-decoder models can reduce the drop to an average of 6% and 2% respectively.
Results suggest a trade-off between performance and robustness, with some models maintaining similar performance while gaining robustness. Our study and adversarial training techniques have been incorporated into an open-source tool for generating camouflaged datasets. However, methodology effectiveness depends on the specific camouflage technique and data encountered, emphasizing the need for continued exploration. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2402_09874 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Camouflage is all you need: Evaluating and Enhancing Language Model Robustness Against Camouflage Adversarial Attacks Huertas-García, Álvaro Martín, Alejandro Huertas-Tato, Javier Camacho, David Computation and Language Adversarial attacks represent a substantial challenge in Natural Language Processing (NLP). This study undertakes a systematic exploration of this challenge in two distinct phases: vulnerability evaluation and resilience enhancement of Transformer-based models under adversarial attacks. In the evaluation phase, we assess the susceptibility of three Transformer configurations, encoder-decoder, encoder-only, and decoder-only setups, to adversarial attacks of escalating complexity across datasets containing offensive language and misinformation. Encoder-only models manifest a 14% and 21% performance drop in offensive language detection and misinformation detection tasks, respectively. Decoder-only models register a 16% decrease in both tasks, while encoder-decoder models exhibit a maximum performance drop of 14% and 26% in the respective tasks. The resilience-enhancement phase employs adversarial training, integrating pre-camouflaged and dynamically altered data. This approach effectively reduces the performance drop in encoder-only models to an average of 5% in offensive language detection and 2% in misinformation detection tasks. Decoder-only models, occasionally exceeding original performance, limit the performance drop to 7% and 2% in the respective tasks. Although not surpassing the original performance, Encoder-decoder models can reduce the drop to an average of 6% and 2% respectively. Results suggest a trade-off between performance and robustness, with some models maintaining similar performance while gaining robustness. Our study and adversarial training techniques have been incorporated into an open-source tool for generating camouflaged datasets. However, methodology effectiveness depends on the specific camouflage technique and data encountered, emphasizing the need for continued exploration. |
| title | Camouflage is all you need: Evaluating and Enhancing Language Model Robustness Against Camouflage Adversarial Attacks |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2402.09874 |