ValCAT: Variable-Length Contextualized Adversarial Transformations Using Encoder-Decoder Language Model

Anonymous

Paper Link: https://openreview.net/forum?id=fE6Md7R_vqA

Paper Type: Long paper (up to eight pages of content + unlimited references and appendices)

Abstract: Adversarial texts help explore vulnerabilities in language models, improve model robustness, and explain their working mechanisms. However, existing word-level attack methods trap in a one-to-one attack pattern, i.e., only a single word can be modified in one transformation round, and they ignore the interactions between several consecutive words. In this paper, we propose ValCAT, a black-box attack framework that misleads the language model by applying variable-length contextualized transformations to the original text. Compared to word-level methods, ValCAT expands the basic units of perturbation from single words to spans composed of multiple consecutive words, enhancing the perturbation capability. Experiments show that our method outperforms state-of-the-art methods in terms of attack success rate, perplexity, and semantic similarity on several classification tasks and inference tasks. The comprehensive human evaluation demonstrates that ValCAT has a significant advantage in ensuring the fluency of the adversarial examples and achieves better semantic consistency. We release the code at https://github.com/linerxliner/ValCAT.

Presentation Mode: This paper will be presented virtually

Virtual Presentation Timezone: UTC+8

Copyright Consent Signature (type Name Or NA If Not Transferrable): Chuyun Deng

Copyright Consent Name And Address: Tsinghua University, Haidian District, Beijing, 100084, P. R. China