RULE: Reliable Multimodal RAG for Factuality in Medical Vision Language Models
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866913549700300800 |
|---|---|
| author | Xia, Peng Zhu, Kangyu Li, Haoran Zhu, Hongtu Li, Yun Li, Gang Zhang, Linjun Yao, Huaxiu |
| author_facet | Xia, Peng Zhu, Kangyu Li, Haoran Zhu, Hongtu Li, Yun Li, Gang Zhang, Linjun Yao, Huaxiu |
| contents | The recent emergence of Medical Large Vision Language Models (Med-LVLMs) has enhanced medical diagnosis. However, current Med-LVLMs frequently encounter factual issues, often generating responses that do not align with established medical facts. Retrieval-Augmented Generation (RAG), which utilizes external knowledge, can improve the factual accuracy of these models but introduces two major challenges. First, limited retrieved contexts might not cover all necessary information, while excessive retrieval can introduce irrelevant and inaccurate references, interfering with the model's generation. Second, in cases where the model originally responds correctly, applying RAG can lead to an over-reliance on retrieved contexts, resulting in incorrect answers. To address these issues, we propose RULE, which consists of two components. First, we introduce a provably effective strategy for controlling factuality risk through the calibrated selection of the number of retrieved contexts. Second, based on samples where over-reliance on retrieved contexts led to errors, we curate a preference dataset to fine-tune the model, balancing its dependence on inherent knowledge and retrieved contexts for generation. We demonstrate the effectiveness of RULE on medical VQA and report generation tasks across three datasets, achieving an average improvement of 47.4% in factual accuracy. We publicly release our benchmark and code in https://github.com/richard-peng-xia/RULE. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2407_05131 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | RULE: Reliable Multimodal RAG for Factuality in Medical Vision Language Models Xia, Peng Zhu, Kangyu Li, Haoran Zhu, Hongtu Li, Yun Li, Gang Zhang, Linjun Yao, Huaxiu Machine Learning Artificial Intelligence Computation and Language Computer Vision and Pattern Recognition Computers and Society The recent emergence of Medical Large Vision Language Models (Med-LVLMs) has enhanced medical diagnosis. However, current Med-LVLMs frequently encounter factual issues, often generating responses that do not align with established medical facts. Retrieval-Augmented Generation (RAG), which utilizes external knowledge, can improve the factual accuracy of these models but introduces two major challenges. First, limited retrieved contexts might not cover all necessary information, while excessive retrieval can introduce irrelevant and inaccurate references, interfering with the model's generation. Second, in cases where the model originally responds correctly, applying RAG can lead to an over-reliance on retrieved contexts, resulting in incorrect answers. To address these issues, we propose RULE, which consists of two components. First, we introduce a provably effective strategy for controlling factuality risk through the calibrated selection of the number of retrieved contexts. Second, based on samples where over-reliance on retrieved contexts led to errors, we curate a preference dataset to fine-tune the model, balancing its dependence on inherent knowledge and retrieved contexts for generation. We demonstrate the effectiveness of RULE on medical VQA and report generation tasks across three datasets, achieving an average improvement of 47.4% in factual accuracy. We publicly release our benchmark and code in https://github.com/richard-peng-xia/RULE. |
| title | RULE: Reliable Multimodal RAG for Factuality in Medical Vision Language Models |
| topic | Machine Learning Artificial Intelligence Computation and Language Computer Vision and Pattern Recognition Computers and Society |
| url | https://arxiv.org/abs/2407.05131 |