arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2203.00249cs.CLcs.AI
收藏

汉语GPT拼音输入法的探索与改造

Exploring and Adapting Chinese GPT to Pinyin Input Method

Singapore Management University(新加坡管理大学);Tencent AI Lab(腾讯人工智能实验室);Zhejiang University(浙江大学)

Minghuan Tan, Yong Dai, Duyu Tang, Zhangyin Feng, Guoping Huang, Jing Jiang, Jiwei Li, Shuming Shi

摘要:

虽然GPT已经成为文本生成任务的事实上的方法,它的应用程序,可编程输入法仍然没有探索。本文首次探索了利用中文GPT进行汉字输入的方法。我们发现,一个冷冻的GPT实现国家的最先进的性能完美的包装。但是,当输入中包含缩写字符串时,性能会急剧下降。一个原因是,一个缩写的汉字可以映射到许多完美的汉字,这些汉字链接到更大数量的汉字。我们用两种策略来缓解这个问题,包括用candidate来丰富上下文,以及优化训练过程以帮助区分同音异义词。为了进一步方便对XML输入法的评估,我们创建了一个由来自15个域的270K实例组成的数据集。结果表明,我们的方法提高了所有领域缩写拼音的性能。模型分析表明,这两种策略都有助于提高性能。

英文摘要:

While GPT has become the de-facto method for text generation tasks, its application to pinyin input method remains unexplored. In this work, we make the first exploration to leverage Chinese GPT for pinyin input method. We find that a frozen GPT achieves state-of-the-art performance on perfect pinyin. However, the performance drops dramatically when the input includes abbreviated pinyin. A reason is that an abbreviated pinyin can be mapped to many perfect pinyin, which links to even larger number of Chinese characters. We mitigate this issue with two strategies, including enriching the context with pinyin and optimizing the training process to help distinguish homophones. To further facilitate the evaluation of pinyin input method, we create a dataset consisting of 270K instances from 15 domains. Results show that our approach improves performance on abbreviated pinyin across all domains. Model analysis demonstrates that both strategies contribute to the performance boost.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑