Examining LLMs' Uncertainty Expression Towards Questions Outside Parametric Knowledge

arXiv:2311.09731v2 Announce Type: replace-cross
Abstract: Can large language models (LLMs) express their uncertainty in situations where they lack sufficient parametric knowledge to generate reasonable responses? This work aims to systematically investigate LLMs' behaviors in such situations, emphasizing the trade-off between honesty and helpfulness. To tackle the challenge of precisely determining LLMs' knowledge gaps, we diagnostically create unanswerable questions containing non-existent concepts or false premises, ensuring that they are outside the LLMs' vast training data. By compiling a benchmark, UnknownBench, which …

abstract arxiv challenge cs.ai cs.cl cs.lg express generate honesty knowledge language language models large language large language models llms parametric questions responses trade trade-off type uncertainty work