# Detecting LLM Hallucination Through Layer-wise Information Deficiency: Analysis of Ambiguous Prompts and Unanswerable Questions

arXiv: https://arxiv.org/abs/2412.10246 · HTML: https://mlsanity.com/paper/2412.10246/

## Key facts

- Cited by 9 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 13 Dec 2024.

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

- [Failure Modes in LLM Systems: A System-Level Taxonomy for Reliable AI Applications](https://mlsanity.com/paper/2511.19933/) (2025)
- [MARAG-R1: Beyond Single Retriever via Reinforcement-Learned Multi-Tool Agentic Retrieval](https://mlsanity.com/paper/2510.27569/) (2025)
- [MultiHaluDet: Multilingual Hallucination Detection via LLM Hidden State Probing](https://mlsanity.com/paper/2605.24919/) (2026)
- [Beyond Surface Statistics: Robust Conformal Prediction for LLMs via Internal Representations](https://mlsanity.com/paper/2604.16217/) (2026)
- [Risk-aware Selective Prompting for Hallucination Mitigation in Large Vision-Language Models](https://mlsanity.com/paper/2605.28123/) (2026)
- [When Answers Stray from Questions: Hallucination Detection via Question-Answer Orthogonal Decomposition](https://mlsanity.com/paper/2605.14449/) (2026)
- [Leveraging Visual Signals for Robust Token-Level Uncertainty in Vision-Language Generation](https://mlsanity.com/paper/2605.27136/) (2026)
- [HalluTracer: Pre-Decoding Truthfulness Prediction via Depth-Averaged Probe-Logit](https://mlsanity.com/paper/2608.16353/) (2026)
- [Vroom-Vroom at SHROOM-Visions: A Multi-Judge Committee for Detecting Hallucinated Spans in Vision-Language Outputs](https://mlsanity.com/paper/2609.17327/) (2026)

## Abstract (arXiv)

Large language models (LLMs) frequently generate confident yet inaccurate responses, introducing significant risks for deployment in safety-critical domains. We present a novel, test-time approach to detecting model hallucination through systematic analysis of information flow across model layers. We target cases when LLMs process inputs with ambiguous or insufficient context. Our investigation reveals that hallucination manifests as usable information deficiencies in inter-layer transmissions. While existing approaches primarily focus on final-layer output analysis, we demonstrate that tracking cross-layer information dynamics (LI) provides robust indicators of model reliability, accounting for both information gain and loss during computation. LI integrates easily with pretrained LLMs without requiring additional training or architectural modifications.