科研速览 · Science Skim继续刷下去 · Keep skimming →
◆ Patterns (New York, N.Y.)2026-08-14

Multimodal integrated knowledge transfer to large language models through preference optimization with biomedical applications.

Zhanliang Wang, Da Wu, Quan M Nguyen, Zhuoran Xu, Kai Wang

原始摘要(英文原文)· Original abstract
The scarcity of high-quality multimodal biomedical data limits the effective fine-tuning of large language models (LLMs) for specialized tasks. We introduce MINT (multimodal integrated knowledge transfer), a framework that aligns unimodal decoder-only models with domain-specific patterns from multimodal biomedical data through preference optimization and that is primarily implemented using the odds ratio preference optimization (ORPO) framework. MINT leverages upstream multimodal machine-learning models to transfer domain expertise to downstream text-only or image-only LLMs, as demonstrated in two applications: (1) rare genetic disease prediction from texts and (2) tissue-type classification using cell nucleus images. In both cases, MINT-based models outperform those enhanced with alternative approaches such as supervised fine-tuning (SFT) and retrieval-augmented generation (RAG), even surpassing much larger foundation models in some scenarios. Our study highlights how MINT effectively grafts the classification strengths of encoder-only models into large decoder-only models, thereby enhancing reasoning abilities and reducing hallucination in biomedical applications.
读原文 · Read the paper ↗

AI 追问PRO

登录后使用 AI 追问

讨论区

登录后参与讨论

相关论文 · Related

Multimodal integrated knowledge transfer to large language models through preference optimization with biomedical applications. — 科研速览 Science Skim