Continued pretraining of Llama-2 on medical data yields MEDITRON-70B, which outperforms GPT-3.5 and Med-PaLM while approaching GPT-4 performance on medical benchmarks.
Comeau, Rezarta Islamaj, Aadit Kapoor, Xin Gao, and Zhiyong Lu
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2023 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
MEDITRON-70B: Scaling Medical Pretraining for Large Language Models
Continued pretraining of Llama-2 on medical data yields MEDITRON-70B, which outperforms GPT-3.5 and Med-PaLM while approaching GPT-4 performance on medical benchmarks.