Unlocking the Power of Large Language Models: The Inaugural AWS Competition

Large language models (LLMs) are revolutionizing the field of artificial intelligence, and a recent competition in Singapore showcased the potential of these models to unlock new possibilities. The inaugural AWS League of Large Language Models competition brought together students from various institutions of higher learning to fine-tune LLMs and demonstrate their capabilities.

Students from Singapore Management University (SMU) and Nanyang Polytechnic (NYP) formed Team CogniSync, which emerged victorious in the competition.

The competition, held in partnership with AI Singapore (AI SG) and the Infocomm Media Development Authority (IMDA), aimed to inspire students to master AI and fine-tune LLMs to unlock their full potential. Participants were taught how to use Amazon Sagemaker Jumpstart to train and refine their LLMs, with the goal of outperforming larger-sized generalized LLMs and other competing teams’ models to answer open-ended questions responsibly.

“Learning how to filter and judge the data sets was our most important takeaway. I’d like to go for more competitions like this to learn more about AI,” said team leader Adhitya Naarayana, an NYP student who is self-taught in AI and ML.

The competition was a resounding success, with 64 teams competing in a three-week long battle royale. In the end, six talented teams remained to take part in the finals at ATxInspire: AI Student Developer Conference 2024. Team CogniSync emerged victorious, winning SGD$1000 worth of Amazon credits. Their winning work was a model that went further than others by using PartyRock to generate data to fine-tune their model.

PartyRock, an Amazon Bedrock playground, is a fun and intuitive generative AI app-building playground that allows users to build and scale apps without coding.

The competition also saw an all-female team, Error404, join the competition after hearing about it from an AI SG student user group. The five Year 1 undergraduates from the National University of Singapore (NUS) consider themselves beginners in AI and ML, but they were pleasantly surprised when they made it to the finals and finished fourth.

“The most enjoyable part of the LO.L was fine-tuning the model, where we curated our own data set to answer the questions,” said Katherine Mia Delacruz Fontanilla, Error404’s leader. “It was difficult since we didn’t have AI or ML experience, but we managed to learn a lot through this process.”

The competition highlighted the importance of LLMs in powering AI and generative AI solutions. Fine-tuning LLMs allows organizations to unlock their full potential to tackle real-world problems. For example, LLMs can be tuned to identify toxic language in a gaming community, detect inappropriate interactions between players, and automatically block the text, ban the player(s), and create a safer gaming environment.

SEA-LION, Southeast Asia’s first LLM family, is culturally relevant, localized, and tailored to the region.

In Singapore, AWS is working with AI Singapore on SEA-LION, Southeast Asia’s first LLM family that is culturally relevant, localized, and tailored to the region. This means SEA-LION is tuned to understand local nuance – for example, “LOL” might be “laughing out loud” in many parts of the world, but in Thailand, they would type “&55555” instead – and be able to accurately understand and respond to users with this cultural nuance.

The competition also highlighted the need for more support in training the future workforce. According to a study by AWS and Access Partnership, 94% of Singapore employers envision their companies becoming AI-driven organizations by 2028, and over four in five say hiring AI-skilled talent is a priority. However, 74% can’t find the talent they need.

“At AWS, we are fully aligned with Singapore’s goal to triple the pool of AI practitioners to 15,000 over five years, and with the National AI Strategy 2.0,” said Elsie Tan, country manager, Worldwide Public Sector, AWS.

The inaugural AWS LO.L competition is a step in the right direction, inspiring students to master AI and fine-tune LLMs to unlock their full potential. With programs like AWS LO.L, Singapore is building next-generation AI talents to advance its National AI 2.0 strategy.

AWS AI Spring is a multifaceted collaboration with the Singapore Government across the public sector, workforce, enterprises, startups, communities, and research and development, to help accelerate the adoption of AI and generative AI in Singapore.