# CausalGaze: Unveiling Hallucinations via Counterfactual Graph Intervention in Large Language Models

arXiv: https://arxiv.org/abs/2604.11087 · HTML: https://mlsanity.com/paper/2604.11087/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 18 of its 42 references that are in the map.
- First posted on arXiv 13 Apr 2026.

## Builds on (most cited first, 18 of 18)

- [TruthfulQA: Measuring How Models Mimic Human Falsehoods](https://mlsanity.com/paper/2109.07958/) (2021)
- [TriviaQA: A Large Scale Distantly Supervised Challenge Dataset for Reading Comprehension](https://mlsanity.com/paper/1705.03551/) (2017)
- [Crowdsourcing Multiple Choice Science Questions](https://mlsanity.com/paper/1707.06209/) (2017)
- [GNNExplainer: Generating Explanations for Graph Neural Networks](https://mlsanity.com/paper/1903.03894/) (2019)
- [Parameterized Explainer for Graph Neural Network](https://mlsanity.com/paper/2011.04573/) (2020)
- [INSIDE: LLMs' Internal States Retain the Power of Hallucination Detection](https://mlsanity.com/paper/2402.03744/) (2024)
- [Out-of-Distribution Detection and Selective Generation for Conditional Language Models](https://mlsanity.com/paper/2209.15558/) (2022)
- [Lookback Lens: Detecting and Mitigating Contextual Hallucinations in Large Language Models Using Only Attention Maps](https://mlsanity.com/paper/2407.07071/) (2024)
- [HaloScope: Harnessing Unlabeled LLM Generations for Hallucination Detection](https://mlsanity.com/paper/2409.17504/) (2024)
- [Hallucination Detection in LLMs Using Spectral Features of Attention Maps](https://mlsanity.com/paper/2502.17598/) (2025)
- [ICR Probe: Tracking Hidden State Dynamics for Reliable Hallucination Detection in LLMs](https://mlsanity.com/paper/2507.16488/) (2025)
- [H-Neurons: On the Existence, Impact, and Origin of Hallucination-Associated Neurons in LLMs](https://mlsanity.com/paper/2512.01797/) (2025)
- [CausalSR: Structural Causal Model-Driven Super-Resolution with Counterfactual Inference](https://mlsanity.com/paper/2501.15852/) (2025)
- [LLM Factoscope: Uncovering LLMs' Factual Discernment through Inner States Analysis](https://mlsanity.com/paper/2312.16374/) (2023)
- [HD-NDEs: Neural Differential Equations for Hallucination Detection in LLMs](https://mlsanity.com/paper/2506.00088/) (2025)
- [FRED: Financial Retrieval-Enhanced Detection and Editing of Hallucinations in Language Models](https://mlsanity.com/paper/2507.20930/) (2025)
- [DAHL: Domain-specific Automated Hallucination Evaluation of Long-Form Text through a Benchmark Dataset in Biomedicine](https://mlsanity.com/paper/2411.09255/) (2024)
- [Addressing Hallucinations in Language Models with Knowledge Graph Embeddings as an Additional Modality](https://mlsanity.com/paper/2411.11531/) (2024)

## Abstract (arXiv)

Despite the groundbreaking advancements made by large language models (LLMs), hallucination remains a critical bottleneck for their deployment in high-stakes domains. Existing classification-based methods mainly rely on static and passive signals from internal states, which often captures the noise and spurious correlations, while overlooking the underlying causal mechanisms. To address this limitation, we shift the paradigm from passive observation to active intervention by introducing CausalGaze, a novel hallucination detection framework based on structural causal models (SCMs). CausalGaze models LLMs' internal states as dynamic causal graphs and employs counterfactual interventions to disentangle causal reasoning paths from incidental noise, thereby enhancing model interpretability. Extensive experiments across four datasets and three widely used LLMs demonstrate the effectiveness of CausalGaze, especially achieving 3.3% improvement in AUROC on the TruthfulQA dataset compared to state-of-the-art baselines.