Summary of Wangchanlion and Wangchanx Mrc Eval, by Wannaphong Phatthiyaphaibun et al.
WangchanLion and WangchanX MRC Eval
by Wannaphong Phatthiyaphaibun, Surapon Nonesung, Patomporn Payoungkhamdee, Peerat Limkonchotiwat, Can Udomcharoenchaikit, Jitkapat Sawatphol, Chompakorn Chaksangchaichot, Ekapol Chuangsuwanich, Sarana Nutanong
First submitted to arxiv on: 24 Mar 2024
Categories
- Main: Computation and Language (cs.CL)
- Secondary: Artificial Intelligence (cs.AI)
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 WangchanLion, an instruction fine-tuned model for Machine Reading Comprehension (MRC) in Thai, is developed by leveraging SEA-LION and a collection of instruction following datasets. The model is publicly released along with training data, code, and final model weights under the Apache-2 license. Experimental studies on two Thai MRC datasets, XQuad and Iapp_wiki_qa_squad, demonstrate WangchanLion’s ability to comprehend context and produce accurate answers in 0-shot and 1-shot settings. The evaluation scheme assesses answer correctness, helpfulness, conciseness, and contextuality, going beyond traditional MRC metrics. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary WangchanLion is a new AI model that can understand Thai texts and provide correct answers to questions about them. It’s like having a super-smart reader that can help us make sense of Thai articles and books. To test its abilities, the researchers used two big datasets of Thai text-based questions and answers. They found that WangchanLion can answer these questions accurately even without seeing the answers before! This is important because it shows the model’s ability to understand the context of the text. |
Keywords
» Artificial intelligence » 1 shot
