Latent Causal Void: Explicit Missing-Context Reconstruction for Misinformation Detection

TL;DR

This paper introduces Latent Causal Void, a method that explicitly reconstructs missing context facts to improve misinformation detection, outperforming previous omission-aware approaches.

Contribution

The paper proposes a retrieval-guided detector that explicitly generates missing facts using large language models and incorporates them into graph reasoning for better detection.

Findings

LCV improves macro-F1 scores by over 2.5 points on bilingual benchmarks.

Explicit missing fact reconstruction enhances omission-aware misinformation detection.

Modeling missing cross-source facts is more effective than just evidence attachment or omission signals.

Abstract

Automatic misinformation detection performs well when deception is visible in what an article explicitly states. However, some misinformation articles remain locally coherent and only become misleading once compared with contemporaneous reports that supply background facts the article omits. We study this omission-relevant setting and observe that current omission-aware approaches typically either attach retrieved context as auxiliary evidence or infer a categorical omission signal, leaving the specific missing fact implicit. We propose \emph{Latent Causal Void} (LCV), a retrieval-guided detector that explicitly reconstructs the missing fact for each target sentence and uses it as a textual cross-source relation in graph reasoning. Concretely, LCV retrieves temporally aligned context articles, asks a frozen instruction-tuned large language model to generate a short missing-context description for each sentence--article pair, and feeds the resulting relation text into a heterograph over target sentences and context articles. On the bilingual benchmark of Sheng et al., LCV improves over the strongest omission-aware baseline by $2.56$ and $2.84$ macro-F1 points on the English and Chinese splits, respectively. The results indicate that modeling the missing cross-source fact itself, rather than only attaching retrieved evidence or predicting an omission signal, is a useful representation for omission-aware misinformation detection.