Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2602.05723 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866908815834742784 |
|---|---|
| author | Yin, Taoye Hu, Haoyuan Fan, Yaxin Chen, Xinhao Wu, Xinya Deng, Kai Zhang, Kezun Wang, Feng |
| author_facet | Yin, Taoye Hu, Haoyuan Fan, Yaxin Chen, Xinhao Wu, Xinya Deng, Kai Zhang, Kezun Wang, Feng |
| contents | In financial Retrieval-Augmented Generation (RAG) systems, models frequently rely on retrieved documents to generate accurate responses due to the time-sensitive nature of the financial domain. While retrieved documents help address knowledge gaps, model-generated responses still suffer from hallucinations that contradict the retrieved information. To mitigate this inconsistency, we propose a Reinforcement Learning framework enhanced with Fine-grained Knowledge Verification (RLFKV). Our method decomposes financial responses into atomic knowledge units and assesses the correctness of each unit to compute the fine-grained faithful reward. This reward offers more precise optimization signals, thereby improving alignment with the retrieved documents. Additionally, to prevent reward hacking (e.g., overly concise replies), we incorporate an informativeness reward that encourages the policy model to retain at least as many knowledge units as the base model. Experiments conducted on the public Financial Data Description (FDD) task and our newly proposed FDD-ANT dataset demonstrate consistent improvements, confirming the effectiveness of our approach. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_05723 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Mitigating Hallucination in Financial Retrieval-Augmented Generation via Fine-Grained Knowledge Verification Yin, Taoye Hu, Haoyuan Fan, Yaxin Chen, Xinhao Wu, Xinya Deng, Kai Zhang, Kezun Wang, Feng Artificial Intelligence In financial Retrieval-Augmented Generation (RAG) systems, models frequently rely on retrieved documents to generate accurate responses due to the time-sensitive nature of the financial domain. While retrieved documents help address knowledge gaps, model-generated responses still suffer from hallucinations that contradict the retrieved information. To mitigate this inconsistency, we propose a Reinforcement Learning framework enhanced with Fine-grained Knowledge Verification (RLFKV). Our method decomposes financial responses into atomic knowledge units and assesses the correctness of each unit to compute the fine-grained faithful reward. This reward offers more precise optimization signals, thereby improving alignment with the retrieved documents. Additionally, to prevent reward hacking (e.g., overly concise replies), we incorporate an informativeness reward that encourages the policy model to retain at least as many knowledge units as the base model. Experiments conducted on the public Financial Data Description (FDD) task and our newly proposed FDD-ANT dataset demonstrate consistent improvements, confirming the effectiveness of our approach. |
| title | Mitigating Hallucination in Financial Retrieval-Augmented Generation via Fine-Grained Knowledge Verification |
| topic | Artificial Intelligence |
| url | https://arxiv.org/abs/2602.05723 |