Factuality evaluation of large language model (LLM) outputs requires decomposing text into discrete "atomic" facts. However, existing definitions of atomicity are underspecified, with empirical results showing high disagreement among annotators, both human and model-based, due to unresolved ambiguity in fact decomposition. We present a visual analytics concept to expose and analyze annotation inconsistencies in fact extraction. By visualizing semantic alignment, granularity and referential dependencies, our approach aims to enable systematic inspection of extracted facts and facilitate convergence through guided revision loops, establishing a more stable foundation for factuality evaluation benchmarks and improving LLM evaluation.
@article{arxiv.2509.01460,
title = {Dissecting Atomic Facts: Visual Analytics for Improving Fact Annotations in Language Model Evaluation},
author = {Manuel Schmidt and Daniel A. Keim and Frederik L. Dennig},
journal= {arXiv preprint arXiv:2509.01460},
year = {2025}
}