Do LLMs estimate uncertainty well in instruction-following?

Large language models (LLMs) could be valuable personal AI agents across various domains, provided they can precisely follow user instructions. However, recent studies have shown significant limitations in LLMs' instruction-following capabilities, raising concerns about their reliability in high-stakes applications. Accurately estimating LLMs' uncertainty in adhering to instructions is critical to mitigating deployment risks. We present, to our knowledge, the first systematic evaluation of uncertainty estimation abilities of LLMs in the context of instructionfollowing. Our study identifies key challenges with existing instruction-following benchmarks, where multiple factors are entangled with uncertainty stemming from instruction-following, complicating the isolation and comparison across methods and models. To address these issues, we introduce a controlled evaluation setup with two benchmark versions of data, enabling comprehensive comparison of uncertainty estimation methods under various conditions. Our findings show that existing uncertainty methods struggle, particularly when models make subtle errors in instruction following. While internal model states provide some improvement, they remain inadequate in more complex scenarios. The insights from our controlled evaluation setups provide crucial understanding of LLMs' limitations and potential for uncertainty estimation in instruction-following tasks, paving the way for more trustworthy AI agents. 1 Work done while at Apple.

In this section, we evaluate LLMs' uncertainty estimation abilities using the IFEval dataset (Zhou et al., 2023), applying six baseline methods across four LLMs. We selected IFEval because it is designed so that a simple and deterministic program can verify whether a response follows the in-