Why is there so little talk about #ChatGPT performance in languages other than English?

Recently, #MT #GPT4 research has started focusing on translation involving (non English) low resource or distant languages

https://arxiv.org/pdf/2303.13780.pdf

This (previous) paper experiments with eighteen different translation directions involving high and low resource languages, as well as non English-centric translations
https://arxiv.org/abs/2302.09210

More research in #LLMs and mid-resource and low-resource languages.
Combining different external models can greatly improve #translation results and practically eliminate certain types of hallucinations.
https://arxiv.org/pdf/2303.16104.pdf

The issue is not purely linguistic. Models in Europe can ensure the quality of the data used for training is representative of the culture and ethics of countries outside the US, inc. privacy
Trombetti: "A two-speed world"
#OpenAI #ChatGPT #LUMI #OpenGPTX
https://www.ft.com/content/563256d4-67b2-465e-b219-52d0f91684ef

More studies on translation involving #lowresource languages.

I am still interested in #midresource language generation with #GPT4

https://twitter.com/_SheilaCastilho/status/1643992845119021062?s=20

There seems to be a "global community of #AI researchers testing the language skills of ChatGPT and its rival chatbots and sounding the alarm about evidence that they are significantly less capable in languages other than English."

References

#ChatGPT

https://www.wired.com/story/chatgpt-non-english-languages-ai-revolution/