Reforest improves anomaly detection with language model insights
ReForest: Interpretability Meets Reasoning in Isolation Forest
Artificial Intelligence
Summary
Anomaly detection tools often spot unusual numbers but miss subtle oddities that depend on the situation. The authors created ReForest, which combines a fast detection method called Isolation Forest with language models that understand context. ReForest turns the detection process into plain language explanations, letting the language model help spot harder anomalies by judging how confident the system is. Tests show this method finds many more true odd cases and explains why it thinks so, which is helpful for important decisions.
What this means in practice
- •For hospital data teams: Improve patient anomaly detection by combining statistical detection with contextual understanding to catch subtle health irregularities.
- •For financial fraud analysts: Enhance fraud detection systems by adding interpretable language-based reasoning to spot context-dependent suspicious transactions.
Authors
Juyoung Seo, Geonwoo Bang, Dongho Kim, Moohong Min
Abstract
Statistical anomaly detection methods like Isolation Forest are efficient but suffer from a critical semantic gap. They identify numerical rarities but fail to understand the contextual meaning of data, often missing subtle, context-dependent anomalies. Conversely, Large Language Models possess strong semantic reasoning capabilities but are prohibitively expensive and slow for large-scale deployment. To bridge this gap, we propose ReForest, a novel framework that synergizes the statistical efficiency of Isolation Forest with the reasoning power of LLMs. ReForest transforms the latent decision paths of Isolation Forest into natural-language explanations, enabling an LLM to refine anomaly detection through an uncertainty-aware dynamic ensemble. This approach allows the model to adaptively weigh statistical and semantic signals based on uncertainty-driven instance-level confidence. Extensive experiments on four real-world datasets (Healthcare, Finance) demonstrate that ReForest significantly outperforms baselines, achieving an average 21% improvement in Recall and a 10% gain in F1-score. Our framework not only improves detection performance but also enhances model transparency by grounding semantic narratives in statistical evidence, providing interpretable rationales essential for high-stakes decision-making.