Researchers from Tsinghua University, TAL AI Lab, and Zhipu.AI have developed MathGLM, a model that challenges the belief that large language models (LLMs) struggle with complex arithmetic tasks. MathGLM can perform a wide range of difficult arithmetic operations and solve math word problems with high accuracy. The model’s mathematical reasoning capabilities have proven superior to those of GPT-4, demonstrating the potential of LLMs in mathematical thinking.
Read more at MarkTechPost…