Mitigating Hallucinations in Large Language Models via Semantic Enrichment of Prompts: Insights from BioBERT and Ontological Integration

被引：0
作者：
Penkov, Stanislav [1 ]
机构：
[1] Sofia Univ St Kliment Ohridski, Sofia, Bulgaria
来源：
PROCEEDINGS OF THE SIXTH INTERNATIONAL CONFERENCE COMPUTATIONAL LINGUISTICS IN BULGARIA, CLIB 2024 | 2024年
关键词：
Large Language Models (LLMs); Semantic Prompt Enrichment; Hallucination Mitigation; Domain-Specific Ontologies; BioBERT Entity Recognition;
D O I：
暂无
中图分类号：
TP18 [人工智能理论];
学科分类号：
081104 ; 0812 ; 0835 ; 1405 ;
摘要：
The advent of Large Language Models (LLMs) has been transformative for natural language processing, yet their tendency to produce "hallucinations"-outputs that are factually incorrect or entirely fabricate-dremains a significant hurdle. This paper introduces a proactive methodology for reducing hallucinations by strategically enriching LLM prompts. This involves identifying key entities and contextual cues from varied domains and integrating this information into the LLM prompts to guide the model towards more accurate and relevant responses. Leveraging examples from BioBERT for biomedical entity recognition and ChEBI for chemical ontology, we illustrate a broader approach that encompasses semantic prompt enrichment as a versatile tool for enhancing LLM output accuracy. By examining the potential of semantic and ontological enrichment in diverse contexts, we aim to present a scalable strategy for improving the reliability of AI-generated content, thereby contributing to the ongoing efforts to refine LLMs for a wide range of applications.
引用
收藏
页码：272 / 276
页数：5
相关论文
共 10 条
  • [1] Brown TB, 2020, Arxiv, DOI [arXiv:2005.14165, 10.48550/arXiv.2005.14165, DOI 10.48550/ARXIV.2005.14165]
  • [2] Feldman P, 2023, Arxiv, DOI [arXiv:2306.06085, 10.48550/ARXIV.2306.06085, DOI 10.48550/ARXIV.2306.06085]
  • [3] Kang Haoqiang, 2023, arXiv
  • [4] Martino A., 2023, P EUR SEM WEB C ESWC
  • [5] Peng BL, 2023, Arxiv, DOI [arXiv:2302.12813, DOI 10.48550/ARXIV.2302.12813, 10.48550/arXiv.2302.12813]
  • [6] Piktus A, 2021, Arxiv, DOI [arXiv:2005.11401, 10.48550/arXiv.2005.11401, DOI 10.48550/ARXIV.2005.11401]
  • [7] Rawte V., 2023, Exploring the Relationship between LLM Hallucinations and Prompt Linguistic Nuances: Readability Formality and Concreteness., DOI [10.48550/arXiv.2309.11064, DOI 10.48550/ARXIV.2309.11064]
  • [8] Vu T, 2023, Arxiv, DOI arXiv:2310.03214
  • [9] Wan FQ, 2024, Arxiv, DOI arXiv:2401.10768
  • [10] Zhang Y, 2023, Arxiv, DOI [arXiv:2309.01219, DOI 10.48550/ARXIV.2309.01219]
1