KScope: A Framework for Characterizing the Knowledge Status of Language Models

Characterizing a large language model's (LLM's) knowledge of a given question is challenging. As a result, prior work has primarily examined LLM behavior under knowledge conflicts, where the model's internal parametric memory contradicts information in the external context. However, this does not fully reflect how well the model knows the answer to the question. In this paper, we first introduce a taxonomy of five knowledge statuses based on the consistency and correctness of LLM knowledge modes. We then propose KScope, a hierarchical framework of statistical tests that progressively refines hypotheses about knowledge modes and characterizes LLM knowledge into one of these five statuses. We apply KScope to nine LLMs across four datasets and systematically establish: (1) Supporting context narrows knowledge gaps across models. (2) Context features related to difficulty, relevance, and familiarity drive successful knowledge updates. (3) LLMs exhibit similar feature preferences when partially correct or conflicted, but diverge sharply when consistently wrong. (4) Context summarization constrained by our feature analysis, together with enhanced credibility, further improves update effectiveness and generalizes across LLMs.

We summarize our contributions 1 in this paper as follows:

• We define a taxonomy of five knowledge statuses based on consistency and correctness, and propose KScope, a hierarchical testing framework to characterize LLM knowledge status.

• We apply KScope to nine LLMs across four datasets, and establish that supporting context substantially narrows knowledge gaps across model sizes and families.

• We identify key context features related to difficulty, relevance, and familiarity that drive successful knowledge updates.

• We reveal how LLM feature importance differs based on parametric knowledge status, showing similarity under conflict but divergence when consistently wrong.

• We validate that constrained context summarization, combined with improved credibility, significantly boosts successful knowledge updates across all statuses and generalizes well.