Saved in:
Bibliographic Details
Main Authors: Liu, Qiang, Chen, Xinlong, Ding, Yue, Song, Bowen, Wang, Weiqiang, Wu, Shu, Wang, Liang
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2501.09997
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911135258640384
author Liu, Qiang
Chen, Xinlong
Ding, Yue
Song, Bowen
Wang, Weiqiang
Wu, Shu
Wang, Liang
author_facet Liu, Qiang
Chen, Xinlong
Ding, Yue
Song, Bowen
Wang, Weiqiang
Wu, Shu
Wang, Liang
contents Hallucination has emerged as a significant barrier to the effective application of Large Language Models (LLMs). In this work, we introduce a novel Attention-Guided SElf-Reflection (AGSER) approach for zero-shot hallucination detection in LLMs. The AGSER method utilizes attention contributions to categorize the input query into attentive and non-attentive queries. Each query is then processed separately through the LLMs, allowing us to compute consistency scores between the generated responses and the original answer. The difference between the two consistency scores serves as a hallucination estimator. In addition to its efficacy in detecting hallucinations, AGSER notably reduces computational overhead, requiring only three passes through the LLM and utilizing two sets of tokens. We have conducted extensive experiments with four widely-used LLMs across three different hallucination benchmarks, demonstrating that our approach significantly outperforms existing methods in zero-shot hallucination detection.
format Preprint
id arxiv_https___arxiv_org_abs_2501_09997
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Attention-guided Self-reflection for Zero-shot Hallucination Detection in Large Language Models
Liu, Qiang
Chen, Xinlong
Ding, Yue
Song, Bowen
Wang, Weiqiang
Wu, Shu
Wang, Liang
Computation and Language
Artificial Intelligence
Hallucination has emerged as a significant barrier to the effective application of Large Language Models (LLMs). In this work, we introduce a novel Attention-Guided SElf-Reflection (AGSER) approach for zero-shot hallucination detection in LLMs. The AGSER method utilizes attention contributions to categorize the input query into attentive and non-attentive queries. Each query is then processed separately through the LLMs, allowing us to compute consistency scores between the generated responses and the original answer. The difference between the two consistency scores serves as a hallucination estimator. In addition to its efficacy in detecting hallucinations, AGSER notably reduces computational overhead, requiring only three passes through the LLM and utilizing two sets of tokens. We have conducted extensive experiments with four widely-used LLMs across three different hallucination benchmarks, demonstrating that our approach significantly outperforms existing methods in zero-shot hallucination detection.
title Attention-guided Self-reflection for Zero-shot Hallucination Detection in Large Language Models
topic Computation and Language
Artificial Intelligence
url https://arxiv.org/abs/2501.09997