Controlled Automatic Task-Specific Synthetic Data Generation for Hallucination Detection

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Xie, Yong, Aggarwal, Karan, Ahmad, Aitzaz, Lau, Stephen
Format: Preprint
Publié: 2024
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866908755137921024
author Xie, Yong
Aggarwal, Karan
Ahmad, Aitzaz
Lau, Stephen
author_facet Xie, Yong
Aggarwal, Karan
Ahmad, Aitzaz
Lau, Stephen
contents We present a novel approach to automatically generate non-trivial task-specific synthetic datasets for hallucination detection. Our approach features a two-step generation-selection pipeline, using hallucination pattern guidance and a language style alignment during generation. Hallucination pattern guidance leverages the most important task-specific hallucination patterns while language style alignment aligns the style of the synthetic dataset with benchmark text. To obtain robust supervised detectors from synthetic datasets, we also adopt a data mixture strategy to improve performance robustness and generalization. Our results on three datasets show that our generated hallucination text is more closely aligned with non-hallucinated text versus baselines, to train hallucination detectors with better generalization. Our hallucination detectors trained on synthetic datasets outperform in-context-learning (ICL)-based detectors by a large margin of 32%. Our extensive experiments confirm the benefits of our approach with cross-task and cross-generator generalization. Our data-mixture-based training further improves the generalization and robustness of hallucination detection.
format Preprint
id arxiv_https___arxiv_org_abs_2410_12278
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Controlled Automatic Task-Specific Synthetic Data Generation for Hallucination Detection
Xie, Yong
Aggarwal, Karan
Ahmad, Aitzaz
Lau, Stephen
Computer Vision and Pattern Recognition
Artificial Intelligence
Computation and Language
68T50
I.2.7
We present a novel approach to automatically generate non-trivial task-specific synthetic datasets for hallucination detection. Our approach features a two-step generation-selection pipeline, using hallucination pattern guidance and a language style alignment during generation. Hallucination pattern guidance leverages the most important task-specific hallucination patterns while language style alignment aligns the style of the synthetic dataset with benchmark text. To obtain robust supervised detectors from synthetic datasets, we also adopt a data mixture strategy to improve performance robustness and generalization. Our results on three datasets show that our generated hallucination text is more closely aligned with non-hallucinated text versus baselines, to train hallucination detectors with better generalization. Our hallucination detectors trained on synthetic datasets outperform in-context-learning (ICL)-based detectors by a large margin of 32%. Our extensive experiments confirm the benefits of our approach with cross-task and cross-generator generalization. Our data-mixture-based training further improves the generalization and robustness of hallucination detection.
title Controlled Automatic Task-Specific Synthetic Data Generation for Hallucination Detection
topic Computer Vision and Pattern Recognition
Artificial Intelligence
Computation and Language
68T50
I.2.7
url https://arxiv.org/abs/2410.12278