Authors: Bambang Parmanto, Bayu Aryoyudanta, Wilbert Soekinto, I Made Agus Setiawan, Yuhan Wang, Haomin Hu, Andi Saptono, Yong K. Choi

Published on: March 11, 2024

Impact Score: 7.6

Arxiv code: Arxiv:2403.06857

Summary

• What is new: Developing a Caregiving Language Model (CaLM) using small foundational models (FMs) that outperforms larger models like GPT-3.5 in certain areas.
• Why this is important: Family caregivers often lack formal training, necessitating a tool to enhance their capacity to provide quality care.
• What the research proposes: A reliable CaLM was developed using small FMs fine-tuned with a caregiving knowledge base to support family caregivers.
• Results: Small FMs with RAG outperformed the larger FM (GPT-3.5) across all metrics, offering a more accessible and efficient solution for caregiving support.

Technical Details

Technological frameworks used: Retrieval Augmented Generation (RAG)

Models used: LLaMA-2, Falcon (7B parameters), GPT-3.5

Data used: Caregiving knowledge base gathered from the Internet

Potential Impact

Elderly care services, caregiver support platforms, healthcare technology firms

Want to implement this idea in a business?

We have generated a startup concept here: Carewise.