Summary of Charpoet: a Chinese Classical Poetry Generation System Based on Token-free Llm, by Chengyue Yu et al.
CharPoet: A Chinese Classical Poetry Generation System Based on Token-free LLM
by Chengyue Yu, Lei Zang, Jiaotuan Wang, Chenyi Zhuang, Jinjie Gu
First submitted to arxiv on: 7 Jan 2024
Categories
- Main: Computation and Language (cs.CL)
- Secondary: Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary This paper proposes CharPoet, a Chinese classical poetry generation system that offers effective control over both format and content. Traditional systems rely on keywords as user inputs, limiting their control over content, while large language models (LLMs) improve content control but often make format errors due to token-by-token generation. The proposed architecture generates in a character-by-character manner, allowing precise control over the number of characters. CharPoet outperforms Jiuge-GPT-2 and GPT-4 in terms of format accuracy, achieving scores above 0.96. In terms of content quality, CharPoet surpasses traditional systems and is comparable to other LLMs. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper creates a new way to generate Chinese classical poetry that lets you control both what it says and how it looks. Usually, these kinds of systems can only understand a few words from the user, but this one uses big language models to get better results. The new system generates one character at a time, which helps it keep track of its formatting. It’s way better than other systems at getting the format right, and its poetry is just as good. |
Keywords
* Artificial intelligence * Gpt * Token
