AI Paper Atlas

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

Bolster Hallucination Detection via Prompt-Guided Data Augmentation

Wenyun Li, Zheng Zhang, Dongmei Jiang, Xiangyuan Lan

AAAI 2026 · Large language models › Language, safety and interpretability · approach: Data and hallucination detection · 1 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
use prompt-guided augmented data with CM score instead of HaloScopeLLaMA-3.1-7B · TruthfulQAAUROC (%): 70.16 -> 73.20Improved
use PALE instead of HaloScopeLLaMA-3.1-7B · TyDi QA-GPAUROC (%): 78.38 -> 85.34Improved
use PALE instead of Lexical SimilarityLLaMA-3.1-7B · CoQAAUROC (%): 74.36 -> 57.94Hurt
use PALE instead of HaloScopeQwen-2.5-7B · TruthfulQAAUROC (%): 73.42 -> 83.68Improved
use PALE instead of HaloScopeOPT-6.7B · CoQAAUROC (%): 67.24 -> 78.64Improved
scale base LLM to 13B/14BPALE · TruthfulQAAUROC (%): 73.20 -> 75.51Improved
train on both truthful and hallucinated data instead of truthful only (ablation)PALE · TruthfulQAAUROC gain: 11.7%Improved
train on both truthful and hallucinated data instead of hallucinated only (ablation)PALE · TruthfulQAAUROC gain: 39.0%Improved
use different LLMs for augmentation (Claude best) (ablation)PALE · TruthfulQAAUROCNo clear effect
vary augmentation prompt template (10 templates) (ablation)PALE · TruthfulQAAUROCNo clear effect
use CM score instead of direct binary classifier (ablation)PALE · Four datasetsAUROCImproved
extract embeddings from middle layer instead of high layers (ablation)PALE · TruthfulQAAUROCImproved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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