Summary of Ffn: a Fine-grained Chinese-english Financial Domain Parallel Corpus, by Yuxin Fu et al.
FFN: a Fine-grained Chinese-English Financial Domain Parallel Corpus
by Yuxin Fu, Shijing Si, Leyi Mai, Xi-ang Li
First submitted to arxiv on: 27 Jun 2024
Categories
- Main: Computation and Language (cs.CL)
- Secondary: Artificial Intelligence (cs.AI); Computational Engineering, Finance, and Science (cs.CE)
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 A novel study probes the effectiveness of Large Language Models (LLMs) in machine translation for the financial domain, leveraging a large-scale Chinese-English parallel corpus called FFN. The dataset comprises 1,013 main text articles and 809 titles from mainstream media websites between 2014 and 2023, manually corrected to ensure accuracy. Two LLMs, ChatGPT and ERNIE-bot, are evaluated using BLEU, TER, and chrF scores, alongside an OpenNMT model trained on the FFN dataset. The study highlights the need for optimizing LLMs in financial translation to ensure quality and accuracy. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary LLMs have greatly improved machine translation, but their performance within the financial domain is largely unknown. Researchers created a large parallel corpus of Chinese-English news articles from 2014 to 2023 to test LLMs’ capabilities. Two models, ChatGPT and ERNIE-bot, were evaluated using special scores, and an OpenNMT model was also trained on the data. The study shows that LLMs need to be fine-tuned for financial translation to work well. |
Keywords
* Artificial intelligence * Bleu * Translation
