WangchanLion and WangchanX MRC Eval
This work addresses the problem of machine reading comprehension for the Thai language, providing an open-source model and datasets to promote research in this domain-specific area.
The researchers developed WangchanLion, an instruction fine-tuned model for Thai machine reading comprehension (MRC), and evaluated it on two Thai MRC datasets (XQuAD and Iapp_wiki_qa_squad) in 0-shot and 1-shot settings, demonstrating its ability to comprehend context and produce faithful answers. They also proposed a new evaluation scheme assessing correctness, helpfulness, conciseness, and contextuality.
This technical report describes the development of WangchanLion, an instruction fine-tuned model focusing on Machine Reading Comprehension (MRC) in the Thai language. Our model is based on SEA-LION and a collection of instruction following datasets. To promote open research and reproducibility, we publicly release all training data, code, and the final model weights under the Apache-2 license. To assess the contextual understanding capability, we conducted extensive experimental studies using two Thai MRC datasets, XQuAD and Iapp_wiki_qa_squad. Experimental results demonstrate the model's ability to comprehend the context and produce an answer faithful to the reference one in 0-shot and 1-shot settings. In addition, our evaluation goes beyond the traditional MRC. We propose a new evaluation scheme assessing the answer's correctness, helpfulness, conciseness, and contextuality. Our code is available publicly at https://github.com/vistec-AI/WangchanLion.