In this paper, we aim to develop an open-source, multilingual language model
for medicine, that the benefits a wider, linguistically diverse audience from
different regions. In general, we present the contribution from the following
aspects: first, for multilingual medical-specific adaptation, we construct a
new multilingual medical corpus, that contains approximately 25.5B tokens
encompassing 6 main languages, termed as MMedC, that enables auto-regressive
training for existing general LLMs. second, to monitor the development of
multilingual LLMs in medicine, we propose a new multilingual medical
multi-choice question-answering benchmark with rationale, termed as MMedBench;
third, we have assessed a number of popular, opensource large language models
(LLMs) on our benchmark, along with those further auto-regressive trained on
MMedC, as a result, our final model, termed as MMedLM 2, with only 7B
parameters, achieves superior performance compared to all other open-source
models, even rivaling GPT-4 on MMedBench. We will make the resources publicly
available, including code, model weights, and datasets.

我们旨在开发一个开源、多语言的医学语言模型，以使更广泛、语言多样的受众从不同地区受益。我们构建了一个新的多语言医学语料库 MMedC，其中包含约 255 亿个标记，涵盖 6 种主要语言，可用于现有通用语言模型的自回归训练。我们还提出了一个新的多语言医学多选题答题基准 MMedBench，带有理由支持的。我们在基准测试中评估了一些流行的开源大型语言模型（LLMs），以及那些在 MMedC 上进一步自回归训练的模型，结果我们的最终模型 MMedLM 2 只有 70 亿个参数，在 MMedBench 上表现出优异的性能，甚至与 GPT-4 不相上下。我们将公开提供这些资源，包括代码、模型权重和数据集。