SavedAdd your journal

Journals / International Journal of 3D Printing Technologies and Digital Industry / 2025 / Cilt: 9 Sayı: 3

ADVANTAGES AND DISADVANTAGES OF RETRIEVAL AUGMENT-ED GENERATION (RAG) IN GENERATIVE AI

Mehmet Taha Toy(ISPARTA UYGULAMALI BİLİMLER ÜNİVERSİTESİ)0009-0000-4220-6224
Ahmet Ali Süzen(Isparta Uygulamalı Bilimler Üniversitesi)0000-0002-5871-1652
Pages
462–475
Publication Date
2025-12-28
🏛️Bibliographic metadata and full text access for this article are provided via DergiPark.
DergiPark Record ↗
✨

AI Summary & Key Takeaways

AI Insight
⚡ TL;DR(Too Long; Didn't Read)

However, significant disadvantages such as system complexity, problems arising from poor quality data and ethical concerns have also been identified.

📌 3 Key Takeaways

🎯
Objective & Hypothesis:In this study, the Retrieval-Augmented Generation (RAG) technology, which is an important devel-opment in the field of artificial intelligence, was examined and its performance was evaluated based on its models.
📊
Key Findings & Data:RAG systems are a new generation technology that helps overcome the limited knowledge of traditional large language models (LLMs), allowing them to use the most up-to-date information and generate more accurate content.
💡
Conclusion & Impact:However, significant disadvantages such as system complexity, problems arising from poor quality data and ethical concerns have also been identified.
🔬Design:Modeling & Simulation
📍Scope:Comprehensive Sample
⚙️Variables:Retrieval Augmented Generation · Large Language Models · Information Access

Abstract

In this study, the Retrieval-Augmented Generation (RAG) technology, which is an important devel-opment in the field of artificial intelligence, was examined and its performance was evaluated based on its models. RAG systems are a new generation technology that helps overcome the limited knowledge of traditional large language models (LLMs), allowing them to use the most up-to-date information and generate more accurate content. The study addresses the basic working mechanism, principles, benefits and technical challenges of RAG technology. Research findings show that RAG systems increase information reliability, reduce the formation of fake findings generated by artificial intelligence, and provide quality results in sectoral applications. However, significant disadvantages such as system complexity, problems arising from poor quality data and ethical concerns have also been identified.

Özet

In this study, the Retrieval-Augmented Generation (RAG) technology, which is an important devel-opment in the field of artificial intelligence, was examined and its performance was evaluated based on its models. RAG systems are a new generation technology that helps overcome the limited knowledge of traditional large language models (LLMs), allowing them to use the most up-to-date information and generate more accurate content. The study addresses the basic working mechanism, principles, benefits and technical challenges of RAG technology. Research findings show that RAG systems increase information reliability, reduce the formation of fake findings generated by artificial intelligence, and provide quality results in sectoral applications. However, significant disadvantages such as system complexity, problems arising from poor quality data and ethical concerns have also been identified.

🔖

Cite This Article

Toy, M. T., Süzen, A. A. (2025). ADVANTAGES AND DISADVANTAGES OF RETRIEVAL AUGMENT-ED GENERATION (RAG) IN GENERATIVE AI. International Journal of 3D Printing Technologies and Digital Industry, Cilt: 9 Sayı: 3, 462–475. https://doi.org/10.46519/ij3dptdi.1757444

📄 Full Text PDF Viewer

📄

PDF görüntüleyici hazırlanıyor...

References (23)

  1. Radford, A., Wu, J., Child, R., Luan, D., Amo-dei, D., & Sutskever, I., “Language models are unsupervised multitask learners”, OpenAI Blog, Vol. 1 Issue 8, 2019.
  2. Brown, T., Mann, B., Ryder, N., Subbiah, M., Kaplan, J. D., Dhariwal, P., Amodei, D., “Lan-guage models are few-shot learners” In 34th Conference on Neural Information Processing Systems (NeurIPS 2020), Vancouver, Canada, Vol. 33, Pages 1877-1901, 2020. 3. Ryckelynck, D., Casenave, F., & Akkari, N. “Error Estimation. In Reduced Order Models and Applications”, Pages 39-52. Springer Nature. doi:10.1007/978-3-031-52764-7_3, 2024.
  3. Lewis, P., Perez, E., Piktus, A., Petroni, F., Karpukhin, V., Goyal, N., Kiela, D., “Retrieval-augmented generation for knowledge-intensive NLP tasks”, Advances in Neural Information Processing Systems, Vol 33, Pages 9459-9474, 2020.
  4. Huang, Y., & Huang, X., “A Survey on Re-trieval-Augmented Text Generation for Large Language Models”, arXiv:2404.10981, 2020.
  5. Gao, Y., Xiong, Y., Gao, X., Jia, K., Pan, J., Bi, Y., Wang, H., “Retrieval-augmented generation for large language models: A survey”, arXiv preprint arXiv:2312.10997. doi:10.48550/arxiv.2312.10997, 2023.
  6. Karpukhin, V., Oguz, B., Min, S., Lewis, P., Wu, L., Edunov, S., Yih, W. T. “Dense passage retrieval for open-domain question answering”, Proceedings of the 2020 Conference on Empiri-cal Methods in Natural Language Processing (EMNLP), Pages 6769-6781, 2020.
  7. Fan, W., Ding, Y., Ning, L., Wang, S., Li, H., Yin, D., Li, Q., “A Survey on RAG Meeting LLMs: Towards Retrieval-Augmented Large Language Models”, Proceedings of the 30th ACM SIGKDD Conference on Knowledge Dis-covery and Data Mining, Vol. 24, Pages 6491-6501, 2024.
  8. Izacard, G., & Grave, E., “Leveraging passage retrieval with generative models for open domain question answering”, Proceedings of the 16th Conference of the European Chapter of the As-sociation for Computational Linguistics: Main Vol. 1282, Pages 874-880, 2021.
  9. Béchard, P., & Ayala, O. M., “Reducing hal-lucination in structured outputs via Retrieval-Augmented Generation”, arXiv preprint doi:10.48550/arxiv.2404.08189, 2024.
  10. Shuster, K., Poff, S., Chen, M., Kiela, D., & Weston, J., “Retrieval augmentation reduces hallucination in conversation”, Findings of the Association for Computational Linguistics: EMNLP, Pages 3784-3803, 2021.
  11. Kang, H., Zhu, Y., Zhong, Y., & Wang, K., Implementing Streaming algorithm and k-means clusters to RAG”. arXiv preprint 2407, 2024. 13. Guu, K., Lee, K., Tung, Z., Pasupat, P., & Chang, M. W., “REALM: Retrieval-augmented language model pre-training” Proceedings of the 37th International Conference on Machine Learning, Vol. 119, Pages 3929-3938, 2020.
  12. Malphedwar, L., Monika, A., Mangesh, D., Tanmay, D., & Raman, G. (2024). Health Diag-nostic Assistant using LLMs. International Jour-nal of Advanced Research in Science, Communi-cation and Technology, Vol. 4 Issue 1, Pages 204-210, 2024.
  13. Ng, T. B., Wu, T.-H., Chen, B. M., Chen, Y., & Kao, B. (2024). A Multi-Stage Prompting and RAG Approach to Generating Legal Analysis in Common Law Systems. Frontiers in Artificial Intelligence and Applications, 385, 190-197. doi:10.3233/faia241278
  14. Tanashchuk, K., Derkach, O., Bazyka, S., & Yamnyuk, B., “Adaptive educational ecosystems using artificial intelligence for forming students' educational trajectories”, Zeszyty Naukowe Wyższej Szkoły Humanitas, Vol 25 Issue 3, Pag-es 97-106, 2024.
  15. Veturi, S., Vaichal, S., Jagadheesh, R. L., Tripto, N. I., & Yan, N., “RAG based Question-Answering for Contextual Response Prediction System”, arXiv preprint arXiv:2409.03708, 2024.
  16. Fleischer, D., Berchansky, M., Wasserblat, M., & Izsak, P., “RAG Foundry: A Framework for Enhancing LLMs for Retrieval Augmented Gen-eration”. arXiv preprint arXiv:2408.02545, 2024.
  17. Zhou, Y., Liu, Y., Li, X., Jin, J. S., Qian, H., Liu, Z., Yu, P. S., “Trustworthiness in Retrieval-Augmented Generation Systems: A Survey. doi:10.48550/arxiv.2409.10102, 2024. 20. Chan, B., Chen, C.-T., Cheng, J.-H., & Huang, H. T., “Don't Do RAG: When Cache-Augmented Generation is All You Need for Knowledge Tasks” arXiv preprint doi:10.48550/arxiv.2412.15605, 2024.
  18. Pickett, M., Hartman, J., Bhowmick, A. K., Alam, R., & Vempaty, A., “Better RAG using Relevant Information Gain”, arXiv preprint doi:10.48550/arxiv.2407.12101, 2024.
  19. Amato, F., Cirillo, E., Fonisto, M., & Moc-cardi, A., “Optimizing Legal Information Access: Federated Search and RAG for Secure AI-Powered Legal Solutions”, 2024 IEEE Interna-tional Conference on Big Data (BigData), Pages 7632-7639, 2024.
  20. Mehrotra, R., Awadallah, A. H., & Yilmaz, E., “Special issue on learning from user interac-tions” Information Retrieval Journal, Vol. 23 Issue 6, Pages 525-527, 2020.
  21. Mao, Q., Zhang, Q., Hao, H., Han, Z., Xu, R., Jiang, W., Yu, P. S., “Privacy-Preserving federat-ed embedding learning for Localized Retrieval-Augmented Generation”, Access address: https://arxiv.org/abs/2504.19101, Access Date, 2025. 25. Wang, S., Zhang, L., Fu, Z., & Mao, Z., “CL-RAG: Bridging the Gap in Retrieval-Augmented Generation with Curriculum Learning”, Access address: https://arxiv.org/abs/2505.10493, Ac-cess Date, 2025.
  22. Ji, Y., Meng, R., Li, Z., & He, D., “Curriculum guided reinforcement learning for efficient multi hop retrieval augmented generation”, Access address: https://arxiv.org/abs/2505.17391, Ac-cess Date, 2025.
  23. Kang, H., Zhu, Y., Zhong, Y., & Wang, K., Implementing Streaming algorithm and k-means clusters to RAG”. arXiv preprint 2407, 2024. 13. Guu, K., Lee, K., Tung, Z., Pasupat, P., & Chang, M. W., “REALM: Retrieval-augmented language model pre-training” Proceedings of the 37th International Conference on Machine Learning, Vol. 119, Pages 3929-3938, 2020.