# LUMINA: Detecting Hallucinations in RAG System with Context-Knowledge Signals

arXiv: https://arxiv.org/abs/2509.21875 · HTML: https://mlsanity.com/paper/2509.21875/

## Key facts

- Cited by 9 papers in the map (data through 14 Oct 2026).
- Placed in Hallucination detection in RAG, one of 15 directions in Retrieval-augmented generation.
- Builds on 27 of its 35 references that are in the map.
- First posted on arXiv 26 Sep 2025.

## Placed in

- Retrieval-augmented generation / [Hallucination detection in RAG](https://mlsanity.com/rag/hallucination-detection-in-rag/)

## Builds on (most cited first, 27 of 27)

- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [Qwen3 Technical Report](https://mlsanity.com/paper/2505.09388/) (2025)
- [Qwen2.5 Technical Report](https://mlsanity.com/paper/2412.15115/) (2024)
- [Retrieval-Augmented Generation for Large Language Models: A Survey](https://mlsanity.com/paper/2312.10997/) (2023)
- [Language Models (Mostly) Know What They Know](https://mlsanity.com/paper/2207.05221/) (2022)
- [A Survey on Hallucination in Large Language Models: Principles, Taxonomy, Challenges, and Open Questions](https://mlsanity.com/paper/2311.05232/) (2023)
- [SelfCheckGPT: Zero-Resource Black-Box Hallucination Detection for Generative Large Language Models](https://mlsanity.com/paper/2303.08896/) (2023)
- [Retrieval Augmentation Reduces Hallucination in Conversation](https://mlsanity.com/paper/2104.07567/) (2021)
- [The Internal State of an LLM Knows When It's Lying](https://mlsanity.com/paper/2304.13734/) (2023)
- [Transformer Feed-Forward Layers Build Predictions by Promoting Concepts in the Vocabulary Space](https://mlsanity.com/paper/2203.14680/) (2022)
- [A Survey on RAG Meeting LLMs: Towards Retrieval-Augmented Large Language Models](https://mlsanity.com/paper/2405.06211/) (2024)
- [Entity-Based Knowledge Conflicts in Question Answering](https://mlsanity.com/paper/2109.05052/) (2021)
- [INSIDE: LLMs' Internal States Retain the Power of Hallucination Detection](https://mlsanity.com/paper/2402.03744/) (2024)
- [RAGTruth: A Hallucination Corpus for Developing Trustworthy Retrieval-Augmented Language Models](https://mlsanity.com/paper/2401.00396/) (2023)
- [Knowledge Conflicts for LLMs: A Survey](https://mlsanity.com/paper/2403.08319/) (2024)
- [Semantic Entropy Probes: Robust and Cheap Hallucination Detection in LLMs](https://mlsanity.com/paper/2406.15927/) (2024)
- [Out-of-Distribution Detection and Selective Generation for Conditional Language Models](https://mlsanity.com/paper/2209.15558/) (2022)
- [Large Language Models with Controllable Working Memory](https://mlsanity.com/paper/2211.05110/) (2022)
- [ReDeEP: Detecting Hallucination in Retrieval-Augmented Generation via Mechanistic Interpretability](https://mlsanity.com/paper/2410.11414/) (2024)
- [Enhancing Uncertainty-Based Hallucination Detection with Stronger Focus](https://mlsanity.com/paper/2311.13230/) (2023)
- [Steer LLM Latents for Hallucination Detection](https://mlsanity.com/paper/2503.01917/) (2025)
- [RefChecker: Reference-based Fine-grained Hallucination Checker and Benchmark for Large Language Models](https://mlsanity.com/paper/2405.14486/) (2024)
- [Hallucination Detection and Hallucination Mitigation: An Investigation](https://mlsanity.com/paper/2401.08358/) (2024)
- [Looking Beyond The Top-1: Transformers Determine Top Tokens In Order](https://mlsanity.com/paper/2410.20210/) (2024)
- [The HalluRAG Dataset: Detecting Closed-Domain Hallucinations in RAG Applications Using an LLM's Internal States](https://mlsanity.com/paper/2412.17056/) (2024)
- ["Lost-in-the-Later": Framework for Quantifying Contextual Grounding in Large Language Models](https://mlsanity.com/paper/2507.05424/) (2025)
- [SEReDeEP: Hallucination Detection in Retrieval-Augmented Models via Semantic Entropy and Context-Parameter Fusion](https://mlsanity.com/paper/2505.07528/) (2025)

## Built on by (most cited first, 9 of 9)

- [How Retrieved Context Shapes Internal Representations in RAG](https://mlsanity.com/paper/2602.20091/) (2026)
- [INTRYGUE: Induction-Aware Entropy Gating for Reliable RAG Uncertainty Estimation](https://mlsanity.com/paper/2603.21607/) (2026)
- [RAGognizer: Hallucination-Aware Fine-Tuning via Detection Head Integration](https://mlsanity.com/paper/2604.15945/) (2026)
- [Detecting Hallucinations in Retrieval-Augmented Generation through Grounding-Aware Sensitivity by Perturbation (GASP)](https://mlsanity.com/paper/2607.04223/) (2026)
- [Position: Avoid Overstretching LLMs for every Enterprise Task](https://mlsanity.com/paper/2605.09365/) (2026)
- [Rethinking Evaluation for LLM Hallucination Detection: A Desiderata, A New RAG-based Benchmark, New Insights](https://mlsanity.com/paper/2605.11330/) (2026)
- [How Much Do Legal RAG Systems Still Hallucinate?](https://mlsanity.com/paper/2608.14210/) (2026)
- [Locating Answer-Correctness Signals in Frozen Large Language Models](https://mlsanity.com/paper/2609.37700/) (2026)
- [Grounding Probes: Generator-Independent Hallucination Detection from Observer Model Hidden States](https://mlsanity.com/paper/2610.04642/) (2026)

## Abstract (arXiv)

Retrieval-Augmented Generation (RAG) aims to mitigate hallucinations in large language models (LLMs) by grounding responses in retrieved documents. Yet, RAG-based LLMs still hallucinate even when provided with correct and sufficient context. A growing line of work suggests that this stems from an imbalance between how models use external context and their internal knowledge, and several approaches have attempted to quantify these signals for hallucination detection. However, existing methods require extensive hyperparameter tuning, limiting their generalizability. We propose LUMINA, a novel framework that detects hallucinations in RAG systems through context--knowledge signals: external context utilization is quantified via distributional distance, while internal knowledge utilization is measured by tracking how predicted tokens evolve across transformer layers. We further introduce a framework for statistically validating these measurements. Experiments on common RAG hallucination benchmarks and four open-source LLMs show that LUMINA achieves consistently high AUROC and AUPRC scores, outperforming prior utilization-based methods by up to +13% AUROC on HalluRAG. Moreover, LUMINA remains robust under relaxed assumptions about retrieval quality and model matching, offering both effectiveness and practicality. LUMINA: https://github.com/deeplearning-wisc/LUMINA