search

HLTCHKUST / chatgpt-evaluation

This respository contains the code for extracting the test samples we used in our paper: "A Multitask, Multilingual, Multimodal Evaluation of ChatGPT on Reasoning, Hallucination, and Interactivity"
76 stars 4 forks source link

Open-sourced multilingual data? #1

Open frankdarkluo opened 11 months ago

frankdarkluo commented 11 months ago

I wonder if the multilingual data used in this paper could be open-sourced or linked?

SamuelCahyawijaya commented 11 months ago

@frankdarkluo: For the multilingual data, we sampled from several datasets, for the sentiment analysis and LID, we use the data from NusaX, while for the MT datasets you can check the detail on the following machine translation script.

Hope it helps!