Context Shapes LLMs Retrieval-Augmented Fact-Checking Effectiveness

TL;DR

This study investigates how context length and evidence placement affect the accuracy of large language models in fact verification, highlighting the importance of prompt structure for improved performance.

Contribution

It provides a comprehensive analysis of the impact of context and evidence placement on LLM-based fact-checking across multiple datasets and models.

Findings

Verification accuracy declines with longer contexts.

Evidence placement near prompt edges improves accuracy.

LLMs possess non-trivial parametric factual knowledge.

Abstract

Large language models (LLMs) show strong reasoning abilities across diverse tasks, yet their performance on extended contexts remains inconsistent. While prior research has emphasized mid-context degradation in question answering, this study examines the impact of context in LLM-based fact verification. Using three datasets (HOVER, FEVEROUS, and ClimateFEVER) and five open-source models accross different parameters sizes (7B, 32B and 70B parameters) and model families (Llama-3.1, Qwen2.5 and Qwen3), we evaluate both parametric factual knowledge and the impact of evidence placement across varying context lengths. We find that LLMs exhibit non-trivial parametric knowledge of factual claims and that their verification accuracy generally declines as context length increases. Similarly to what has been shown in previous works, in-context evidence placement plays a critical role with accuracy being consistently higher when relevant evidence appears near the beginning or end of the prompt and lower when placed mid-context. These results underscore the importance of prompt structure in retrieval-augmented fact-checking systems.