PhoGPT opens up many creative applications

At the Artificial Intelligence Day 2023 event with the theme "AI - redesigning reality" taking place on December 5 and 6, VinAI Artificial Intelligence Research and Application Company ( Vingroup Corporation) announced an open source research project on a large language model for Vietnamese, PhoGPT.

PhoGPT is an open-source project instead of a proprietary software like ChatGPT of OpenAI. Because it is open-source, there is no commercial limitation, all parties can use PhoGPT to develop their own applications, including application units for commercial purposes. This means that this is like a platform providing the community developing applications related to AI technology in the country.

AI Day 2023 takes place over 2 days (December 5, 6) with the participation of more than 30 experts, professors and leading scientists in the field of artificial intelligence from all over the world.

According to Dr. Bui Hai Hung, General Director of VinAI Artificial Intelligence Research and Application Company, the limitations of Vietnamese models have proven that these models have not achieved optimal performance and lack an open source code. Therefore, one of the urgent tasks for the AI community in general, and the natural language processing (NLP) community in particular, is to build a new, more powerful model capable of processing Vietnamese language with high accuracy and performance.

AI experts say that with a big data language model with 7.5 billion parameters, built on the Transformer decoding platform, this model is trained from scratch, using the most advanced techniques available such as Flash Attention mechanism, AliBi context length extrapolation...

These techniques not only help the model gain a deeper understanding of context, but also enhance PhoGPT’s natural dialogue and interaction capabilities. This makes the model a versatile and multi-tasking tool, capable of meeting a wide range of users’ linguistic needs.

Dr. Bui Hai Hung added that PhoGPT was developed by the company from the beginning, independent of all other models in the world . With the open source model, the community in Vietnam can use and improve it better. Making PhoGPT source code public and available to users helps create an environment where the user community can develop customized and unique applications.

One of the goals of open source is to lay a foundation so that people do not have to spend time redoing, and units can develop more large language models such as PhoGPT. This will help society have a quality open source community for large Vietnamese language models, creating a good effect so that many companies can participate and apply them in a certain field. With PhoGPT, VinAI Artificial Intelligence Research and Application Company said that it will have a plan to research and develop applications for individual users and a package of specialized support solutions for businesses in Vietnamese in fields such as healthcare, education, etc.

PhoGPT has laid the first foundations for the development of high-performance Vietnamese language models, as a basis for developing practical, effective applications, in line with the Government's AI development strategy to 2030.

BA TAN

Source