Zero-shot sampling of adversarial entities in biomedical question answering

arXiv:2402.10527v1 Announce Type: cross
Abstract: The increasing depth of parametric domain knowledge in large language models (LLMs) is fueling their rapid deployment in real-world applications. In high-stakes and knowledge-intensive tasks, understanding model vulnerabilities is essential for quantifying the trustworthiness of model predictions and regulating their use. The recent discovery of named entities as adversarial examples in natural language processing tasks raises questions about their potential guises in other settings. Here, we propose a powerscaled distance-weighted sampling scheme in embedding space …

adversarial applications arxiv biomedical cs.cl cs.cr deployment discovery domain entities high knowledge language language models large llms predictions question rapid real stat.ap trustworthiness understanding vulnerabilities world