Talking Nonsense: Probing Large Language Models' Understanding of Adversarial Gibberish Inputs

arXiv:2404.17120v1 Announce Type: cross
Abstract: Large language models (LLMs) exhibit excellent ability to understand human languages, but do they also understand their own language that appears gibberish to us? In this work we delve into this question, aiming to uncover the mechanisms underlying such behavior in LLMs. We employ the Greedy Coordinate Gradient optimizer to craft prompts that compel LLMs to generate coherent responses from seemingly nonsensical inputs. We call these inputs LM Babel and this work systematically studies the …

abstract adversarial arxiv behavior cs.ai cs.cl cs.lg human inputs language language models languages large language large language models llms question type understanding work