StudyArena 于 2026 年 8 月 24 日发布了基于 6,851 次符合条件的学生盲测投票的 AI 模型大学论文写作排名 1。在大学论文写作任务中，Gemini 以 39.6% 的选择率位居榜首，领先于获得 31.8% 选择率的 Claude 以及获得 29.2% 选择率的 ChatGPT/OpenAI 1。
在具体的细分任务中，各模型展现出不同的专长 1。Gemini 在写作反馈方面以 41.7% 的选择率领先，Claude 在作业规划方面以 43.2% 占据优势，而 GPT 则在研究工作方面以 39.3% 领先 1。此外，被选中的回答平均比替代回答长 37%，且最长回答在决定性写作比较中的获胜率为 47.7% 1。
当前竞技场包含的模型版本为 GPT-5.6 Sol、Claude Opus 5 和 Gemini 3.1 Pro 1。文章建议将 Gemini 作为论文编辑工具而非代写工具使用 1。
On August 24, 2026, StudyArena released a blind test ranking of AI models for university essay writing, based on an analysis of data from August 2026 1. The analysis is based on 6,851 eligible blind test votes from students 1. The results reveal that Gemini leads university essay writing tasks with a 39.6% selection rate 1. Claude follows with a 31.8% selection rate, while ChatGPT/OpenAI trails at 29.2% 1. The current arena includes GPT-5.6 Sol, Claude Opus 5, and Gemini 3.1 Pro 1.
The report points out that each model has its own expertise in different tasks 1. Specifically, Gemini leads in writing feedback with a 41.7% selection rate, Claude excels in assignment planning at 43.2%, and GPT takes the lead in research work with 39.3% 1. Additionally, the study notes that the selected answers were, on average, 37% longer than the alternative responses 1. Furthermore, the longest response won in 47.7% of the decisive writing comparisons 1. Ultimately, the report suggests using Gemini as an essay editing tool rather than a ghostwriting tool 1.
评论
还没有评论，欢迎留下第一条。