@inproceedings{13db0d5d8ee1490d8da4eabf79c70ebc,
title = "UHDF: Hallucination Detection Using Open Source Models Beyond Close Source Models Methods",
abstract = "With the emergence of multimodal large models, the problem of hallucination has been plaguing their development and deployment. How to reliably detect the presence of hallucinations in mLLMs has become an important issue. We propose UHDF, which replaces the closed-source models in it with open-source models by improving UniHD [15], and dramatically outperforms it. By optimizing the external information used in UniHD and achieving decoupling between different external information sources, we minimize the hallucinations introduced in pipeline, and thus improve the effectiveness of hallucinations detection. UHDF using the open-source model outperforms UniHD using the closed-source model (GPT-4v), achieving 86.6\% (dev set)/85.3\% (test set) on MacroF1 and achieved the first place in NLPCC2024 Shared Task 10 Track1 (Open Source). Our code and models are available at https://github.com/codetalker125/UHDF.",
keywords = "external information, hallucination detection, mLLM",
author = "Dongxu Liu and Bufan Xu and Zhilong Zhao and Bing Xu and Muyun Yang",
note = "Publisher Copyright: {\textcopyright} The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.; 13th CCF International Conference on Natural Language Processing and Chinese Computing, NLPCC 2024 ; Conference date: 01-11-2024 Through 03-11-2024",
year = "2025",
doi = "10.1007/978-981-97-9443-0\_34",
language = "英语",
isbn = "9789819794423",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
publisher = "Springer Science and Business Media Deutschland GmbH",
pages = "389--399",
editor = "Wong, \{Derek F.\} and Zhongyu Wei and Muyun Yang",
booktitle = "Natural Language Processing and Chinese Computing - 13th National CCF Conference, NLPCC 2024, Proceedings",
address = "德国",
}