With some great advances being made in the field of artificial intelligence, natural language systems are advancing rapidly. Language Large Models (LLMs) are getting significantly better and more popular with each upgrade and innovation. A new feature or mod is added almost daily, which allows LLM to work in different applications in almost every field. LLM is everywhere, from machine translation and text summarization to sentiment analysis and question answering.
The open source community has made some notable progress in developing chat-based LLMs, but mostly in English. Less emphasis has been placed on developing a similar type of multilingual chat capability in LLM. To address this, SambaNova, a software company focused on generative AI solutions, has introduced an open source, multilingual LLM conversation called BLOOMChat. Developed in collaboration with Together, an open, scalable and decentralized AI cloud, BLOOMChat is a 176 billion variables multilingual LLM chat built on the BLOOM model.
The BLOOM model has the ability to generate text in 46 natural languages and 13 programming languages. For languages like Spanish, French, and Arabic, BLOOM represents the first language model ever created with over 100 billion parameters. BLOOM was developed by BigScience, an international collaboration of more than 1,000 researchers. By tuning BLOOM to open conversation and alignment datasets from projects such as OpenChatKit, Dolly 2.0, and OASST1, BLOOM’s core capabilities have been extended to the chat domain.
To develop multilingual chat, LLM, BLOOMChat, SambaNova, and Together used SambaNova DataScale systems that use SambaNova’s unique reconfigurable data flow architecture for the training process. Synthetic conversation data and human written samples were combined to create BLOOMChat. A large synthetic dataset called OpenChatKit was used as the basis for the chat function, and high-quality human-generated datasets such as Dolly 2.0 and OASST1 were used to greatly improve performance. The code and scripts used to set the help on the OpenChatKit and Dolly-v2 datasets are provided on SambaNova’s GitHub.
In human assessments conducted across six languages, BLOOMChat responses were preferred over GPT-4 responses 45.25% of the time. Compared to four other open source chat alignment models with the same six languages, BLOOMChat responses ranked as the best 65.92% of the time. This achievement successfully bridges the multilingual chat capability gap in the open source market. In the WMT localization test, BLOOMChat performed better than additional iterations of the BLOOM model as well as popular open source chat models.
BLOOMChat, like other LLMs, has limitations. It may result in incorrect or factually irrelevant information or it may change languages by mistake. He can even repeat phrases, has limited coding or math abilities, and sometimes produces toxic content. More research is working to address these challenges and ensure better usability.
In conclusion, BLOOMChat builds on the extensive work of the open source community and is a great addition to the list of some very useful multilingual LLMs. Released under an open source license, SambaNova and Together aim to expand access to advanced multilingual chat capabilities and encourage further innovation in the AI research community.
scan the project And Reference article. Don’t forget to join 21k+ML Sub RedditAnd discord channelAnd And Email newsletter, where we share the latest AI research news, cool AI projects, and more. If you have any questions regarding the above article or if we’ve missed anything, feel free to email us at [email protected]
🚀 Check out 100’s AI Tools in the AI Tools Club
Tania Malhotra is a final year from University of Petroleum and Energy Studies, Dehradun, pursuing a BTech in Computer Science Engineering with a specialization in Artificial Intelligence and Machine Learning.
She is passionate about data science and has good analytical and critical thinking, along with a keen interest in acquiring new skills, leading groups, and managing work in an organized manner.
“Unapologetic communicator. Wannabe web lover. Friendly travel scholar. Problem solver. Amateur social mediaholic.”