New Persian Medical LLM Family 'Gaokerena' Released

Researchers have introduced Gaokerena, a new family of small Persian medical language models designed for consumer-grade hardware. The models include Gaokerena-V, trained on a 90-million-token Persian medical corpus, and Gaokerena-R, which incorporates Chain-of-Thought and RLAIF for improved clinical reasoning. Gaokerena-R achieved a higher benchmark score than Gaokerena-V despite using a smaller dataset. Both models feature uncertainty heads to predict response confidence, though further research is needed before clinical deployment. AI

IMPACT This research addresses the gap in low-resource language medical AI, potentially enabling wider access to AI-powered healthcare tools in Persian-speaking regions.

RANK_REASON The cluster contains an academic paper detailing the creation and evaluation of new language models. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.CL →

• Gaokerena
• Gaokerena-R
• Gaokerena-V
• Hugging Face
• Massive Multitask Language Understanding
• Persian
• Reinforcement Learning with AI Feedback

• model release
• paper
• other

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →