RAG for Vehicle Crash Collision Safety Tests

With the increasing interest in leveraging large language models (LLMs) for organizational applications, particularly in industrial management and research, the need for fine-tuning these models to domain-specific information has grown significantly. In collaboration with Hyundai Motor Company, this project introduces a Retrieval-Augmented Generation (RAG) model fine-tuned for car crash safety data. The project utilizes LLMs to extract and interpret domain-specific information presented in a multimodal format. Additionally, we explore retrieval and encoder models capable of effective communication in a non-English context, specifically in Korean.