AI Paper Atlas

Research map / Large language models / Language, safety and interpretability

Lyapunov Probes for Hallucination Detection in Large Foundation Models

Bozhi Luan, Gen Li, Yalan Qin, Jifeng Guo, Yun Zhou, Faguo Wu, Hongwei Zheng, Wenjun Wu, Zhaoxin Fan

CVPR 2026 · Large language models › Language, safety and interpretability · approach: Data and hallucination detection · 3 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add Lyapunov constraint loss (ablation)Lyapunov Probe · TriviaQAAUPRC: Qwen-3-4B 74.19 -> 79.47; Llama-2-7B 78.13 -> 83.09Improved
use multi-layer hidden states instead of single layer (ablation)Lyapunov Probe · TriviaQAAUPRC: Llama-2-7B 77.34 -> 83.09; Falcon-7B 60.50 -> 65.52Improved
add perturbation data in training (ablation)Lyapunov Probe · TriviaQAAUPRC: Llama-3-8B 82.35 -> 86.46; Qwen-3-4B 79.92 -> 79.47Mixed
add two-stage training (ablation)Lyapunov Probe · TriviaQAAUPRC: Llama-3-8B 84.80 -> 86.46; Qwen-3-4B 80.58 -> 79.47Mixed
replace standard probe with Lyapunov ProbeLlama-3-8B · TriviaQAAUPRC: 78.82 -> 86.46Improved
replace standard probe with Lyapunov ProbeQwen-3-4B · PopQAAUPRC: 64.41 -> 64.02No clear effect
replace standard probe with Lyapunov ProbeLlama-2-7B · MMLUAUPRC: 33.60 -> 33.79No clear effect
replace standard probe with Lyapunov ProbeLLaVA-1.5-7B · VizWiz-VQAAUPRC: 77.02 -> 83.18Improved
replace standard probe with Lyapunov ProbeLLaVA-1.5-7B · POPEAUPRC: 98.08 -> 99.13No clear effect
train probe on TriviaQA and test on other datasets instead of in-domainLyapunov Probe · CoQA (cross-domain, Llama-3-8B)AUPRC: 81.28 -> 67.65Hurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy