LimeAttack: Local Explainable Method for Textual Hard-Label Adversarial Attack
This addresses a more realistic and challenging security vulnerability in NLP models, though it is incremental as it builds on existing hard-label attack methods.
The paper tackles the problem of generating adversarial examples for NLP models under a hard-label attack setting, where only discrete prediction labels are available, and shows that LimeAttack achieves better attacking performance than existing methods under the same query budget, with results indicating adversarial examples remain a threat to large language models.
Natural language processing models are vulnerable to adversarial examples. Previous textual adversarial attacks adopt gradients or confidence scores to calculate word importance ranking and generate adversarial examples. However, this information is unavailable in the real world. Therefore, we focus on a more realistic and challenging setting, named hard-label attack, in which the attacker can only query the model and obtain a discrete prediction label. Existing hard-label attack algorithms tend to initialize adversarial examples by random substitution and then utilize complex heuristic algorithms to optimize the adversarial perturbation. These methods require a lot of model queries and the attack success rate is restricted by adversary initialization. In this paper, we propose a novel hard-label attack algorithm named LimeAttack, which leverages a local explainable method to approximate word importance ranking, and then adopts beam search to find the optimal solution. Extensive experiments show that LimeAttack achieves the better attacking performance compared with existing hard-label attack under the same query budget. In addition, we evaluate the effectiveness of LimeAttack on large language models, and results indicate that adversarial examples remain a significant threat to large language models. The adversarial examples crafted by LimeAttack are highly transferable and effectively improve model robustness in adversarial training.