LBA: Textual Hard-Label Adversarial Attack under Low Query Budgets

Jul 17, 2026

A new sampling-based method called LBA is proposed for generating high-quality adversarial texts in hard-label scenarios with low query budgets. LBA constructs an approximate distribution of adversarial examples by integrating prior and posterior knowledge, enabling more effective sampling. Experiments across six language models and four datasets show that LBA outperforms state-of-the-art baselines on all evaluation metrics and produces more semantically preserved adversarial texts according to LLM-based assessments.

Why it matters: Improving the efficiency and quality of adversarial attacks is important for robustly evaluating and strengthening language models against real-world threats.

Full story at: arXiv Computation and Language ↗