Keywords
How to Cite
Abstract
This thesis analyzes the types of hallucinations that occur in large language models, their underlying causes, methods for detecting them, and the main approaches to mitigating them. The methods are classified according to the data, retrieval, generation, and verification stages, and their applicability to low-resource languages such as Uzbek is evaluated.
References
1. Ji Z., Lee N., Frieske R. va boshq. Survey of Hallucination in Natural Language Generation // ACM Computing Surveys. – 2023. – Vol. 55, No. 12. – P. 1–38.
2. Huang L., Yu W., Ma W. va boshq. A Survey on Hallucination in Large Language Models: Principles, Taxonomy, Challenges, and Open Questions // ACM Transactions on Information Systems. – 2025. – Vol. 43, No. 2.
3. Li Y., Fu X., Verma G., Buitelaar P., Liu M. Mitigating Hallucination in Large Language Models (LLMs): An Application-Oriented Survey on RAG, Reasoning, and Agentic Systems // arXiv preprint arXiv:2510.24476. – 2025.
4. Lewis P., Perez E., Piktus A. va boshq. Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks // Advances in Neural Information Processing Systems (NeurIPS). – 2020. – Vol. 33. – P. 9459–9474.
5. Hossain I., Banik M., Al Farid F., Uddin J., Karim H.A. Advancing Large Language Models for Low-Resource Languages: A Systematic Review of Pretraining, Adaptation, and Ethical Challenges // Computer Modeling in Engineering & Sciences. – 2026. – Vol. 148, No. 1.
6. Min S., Krishna K., Lyu X. va boshq. FActScore: Fine-grained Atomic Evaluation of Factual Precision in Long Form Text Generation // Proceedings of EMNLP. – 2023. – P. 12076–12100.
7. Manakul P., Liusie A., Gales M.J.F. SelfCheckGPT: Zero-Resource Black-Box Hallucination Detection for Generative Large Language Models // Proceedings of EMNLP. – 2023. – P. 9004–9017.
8. Es S., James J., Espinosa-Anke L., Schockaert S. RAGAS: Automated Evaluation of Retrieval Augmented Generation // Proceedings of EACL: System Demonstrations. – 2024. – P. 150–158.