Agentic Retrieval-Augmented Generation (RAG) Framework with Quadruple-Based Reasoning and Reinforcement Learning (RL) Optimization

Retrieval-Augmented Generation (RAG) has emerged as an effective technique to reduce hallucinations in large language models (LLMs) but they follow a static retrieve-then-generate pipeline. This process is insufficient for complex financial question answering system that require multi-step reasoning, numerical precision, and factual verification. Therefore, in this research, we proposed an RL-Driven Agentic Multi-HyDE RAG framework designed to improve factual correctness and informativeness through structured reasoning and reinforcement learning optimization. The proposed methodology comprises of six major components: query diversification, hypothetical answer generation (HyDE), dense embedding-based retrieval, quadruple-based atomic knowledge representation, reinforcement learning-based evaluation, and tool-augmented refinement. Experimental evaluation on financial queries using Sentence Transformers, FAISS, and Mistral-7B-Instruct demonstrates that the framework achieves high factual alignment (faithfulness score = 1.0) while maintaining informativeness, without unnecessary calling of the external tools. The results indicate that integrating agentic reasoning, structured knowledge extraction, and reinforcement learning significantly overcomes hallucinations and improves reliability. The proposed architecture provides a scalable and robust solution for high-stakes financial question answering systems.

Retrieval-Augmented Generation (RAG); Agentic AI; Reinforcement Learning; Hypothetical Document Embeddings (HyDE); Financial Question Answering

1. Chen, M., Li, T., Sun, H., Zhou, Y., Zhu, C., Wang, H., Pan, J. Z., Zhang, W., Chen, H., Yang, F., Zhou, Z., & Chen, W. (2025). ReSearch: Learning to reason with search for LLMs via reinforcement learning. [Google Scholar] [Crossref]

2. Eibich, M., Nagpal, S., & Fred-Ojala, A. (2024). ARAGOG: Advanced RAG output grading. arXiv preprint arXiv:2404.01037. [Google Scholar] [Crossref]

3. Fang, F., Bai, Y., Ni, S., Yang, M., Chen, X., & Xu, R. (2024). Enhancing noise robustness of retrievalaugmented language models with adaptive adversarial training. Proceedings of the ACL, 10028–10039. [Google Scholar] [Crossref]

4. Gao, L., Ma, X., Lin, J., & Callan, J. (2023). Precise zero-shot dense retrieval without relevance labels. Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics, 1762–1777. [Google Scholar] [Crossref]

5. Guo, D., Yang, D., Zhang, H., et al. (2025). DeepSeek-R1: Incentivizing reasoning capability in LLMs via reinforcement learning. [Google Scholar] [Crossref]

6. Guan, X., Zeng, J., Meng, F., Xin, C., Lu, Y., Lin, H., Han, X., Sun, L., & Zhou, J. (2025). DeepRAG: Thinking to retrieve step by step for large language models. [Google Scholar] [Crossref]

7. Hong, G., Kim, J., Kang, J., Myaeng, S.-H., & Whang, J. (2024). Why so gullible? Enhancing the robustness of retrieval-augmented models against counterfactual noise. Findings of NAACL 2024, 2474– 2495. [Google Scholar] [Crossref]

8. Jaech, A., Kalai, A., Lerer, A., et al. (2024). OpenAI o1 system card. [Google Scholar] [Crossref]

9. Jiang, Z., Xu, F., Gao, L., Sun, Z., Liu, Q., Dwivedi-Yu, J., Yang, Y., Callan, J., & Neubig, G. (2023). [Google Scholar] [Crossref]

10. Active retrieval augmented generation. Proceedings of EMNLP, 7969–7992. [Google Scholar] [Crossref]

11. LangChain. (2023). Query transformations. [Google Scholar] [Crossref]

12. Li, X., Dong, G., Jin, J., Zhang, Y., Zhou, Y., Zhu, Y., Zhang, P., & Dou, Z. (2025). Search-o1: Agentic search-enhanced large reasoning models. [Google Scholar] [Crossref]

13. Li, Y., Luo, Q., Li, X., Li, B., Cheng, Q., Wang, B., Zheng, Y., Wang, Y., Yin, Z., & Qiu, X. (2025). R3RAG: Learning step-by-step reasoning and retrieval for LLMs via reinforcement learning. [Google Scholar] [Crossref]

14. Ma, X., Gong, Y., He, P., Zhao, H., & Duan, N. (2023). Query rewriting in retrieval-augmented large language models. Proceedings of EMNLP, 5303–5315. [Google Scholar] [Crossref]

15. Song, H., Jiang, J., Min, Y., Chen, J., Chen, Z., Zhao, W. X., Fang, L., & Wen, J.-R. (2025). R1-Searcher: Incentivizing the search capability in LLMs via reinforcement learning. [Google Scholar] [Crossref]

16. Yu, T., Zhang, S., & Feng, Y. (2024). Auto-RAG: Autonomous retrieval-augmented generation for large language models. [Google Scholar] [Crossref]

17. Zhang, T., Li, K., Luo, H., Wu, X., Glass, J. R., & Meng, H. M. (2024). Adaptive query rewriting: Aligning rewriters through marginal probability of conversational answers. Proceedings of EMNLP, 13444–13461. [Google Scholar] [Crossref]

• A Machine Learning Model for Predicting the Risk of Developing Diabetes - T2DM Using Real-World Data from Kilifi, Kenya
• AI-Powered Facial Recognition Attendance System Using Deep Learning and Computer Vision
• A Comprehensive Review on Brain Tumour Segmentation Using Deep Learning Approach
• A Scalable Retrieval-Augmented Generation Pipeline for Domain-Specific Knowledge Applications
• Predictive Maintenance in Semiconductor Manufacturing Using Machine Learning on Imbalanced Dataset