Parametric Knowledge is Not All You Need: Toward Honest Large Language Models via Retrieval of Pretraining Data
This addresses the issue of hallucination in LLMs for users relying on accurate information, though it is incremental as it builds on existing honesty methods.
The paper tackles the problem of large language models (LLMs) generating incorrect responses due to unawareness of their knowledge boundaries, and proposes a robust evaluation benchmark using Pythia's open pretraining data, along with a method to improve honesty by leveraging this data.
Large language models (LLMs) are highly capable of answering questions, but they are often unaware of their own knowledge boundary, i.e., knowing what they know and what they don't know. As a result, they can generate factually incorrect responses on topics they do not have enough knowledge of, commonly known as hallucination. Rather than hallucinating, a language model should be more honest and respond with "I don't know" when it does not have enough knowledge about a topic. Many methods have been proposed to improve LLM honesty, but their evaluations lack robustness, as they do not take into account the knowledge that the LLM has ingested during its pretraining. In this paper, we propose a more robust evaluation benchmark dataset for LLM honesty by utilizing Pythia, a truly open LLM with publicly available pretraining data. In addition, we also propose a novel method for harnessing the pretraining data to build a more honest LLM.