Fig. 1: Comparison of poetry generation between Xunzi-Yayun-R1 and DeepSeek Models.
From: Large language models learning to write rhyming Tang poetry A Xunzi Yayun R1 case study
This chart shows a comparison between the Xunzi-Yayun-R1 model and the DeepSeek model on the task of Tang poetry generation. On the left is the Xunzi-Yayun-R1 model, and on the right is the DeepSeek model. Both were tasked with answering the same classical poetry creation prompts, including composing poems based on specific rhymes and themes.
