The One About Making AI More Singaporean
What does it take to make AI truly understand Singapore?
While LLMs excel on global benchmarks today, adapting them to Singapore's unique language, culture, and social context remains an open research challenge. Join researchers from A*STAR, SMU, and NUS as they share their research in multilingual language models, cultural alignment, controllable Singlish generation and retrieval-augmented generation. Explore how these techniques are enabling AI systems to better understand and generate language for Singapore.
More About the Sharings
Weihua Zheng (Senior Research Engineer, A*STAR) will share on "Understanding Local Context with AI"
Understanding local language, cultural references, and social context remains a significant challenge for today's large language models. Drawing on his research in multilingual language models, Weihua will explore how AI systems can better adapt to Singapore's unique multilingual and multicultural environment. He will also discuss how locally grounded datasets, cultural evaluation, and alignment techniques can help build AI systems that are more relevant, reliable, and responsive to local users. (Technical Level: 100)
Liang Jinggui (PhD Candidate, SMU) will share on "Teaching AI to Speak Singlish"
Converting between Standard English and Singlish involves more than swapping words. It requires understanding syntax, code-switching, pragmatics, and other linguistic features that define Singapore's unique way of speaking. Jinggui will share how his team developed an explainable, fine-grained approach to Singlish style transfer, allowing language models to control and explain individual stylistic changes rather than treating them as a black box. (Technical Level: 200)
More About the Speakers
Weihua Zheng is currently a Senior Research Engineer at A*STAR and a PhD student at the Singapore University of Technology and Design (SUTD). His research interests include machine translation, enhancing the capabilities of multilingual large language models, and aligning large language models with human preferences and cultural knowledge. He has served as a key technical developer for Singapore Translate Together, CATOS, and several national-level research projects in Singapore. He has extensive experience in advancing the application of LLMs to multilingual and multicultural tasks.
Liang Jinggui is a PhD candidate in Computer Science at the School of Computing and Information Systems, Singapore Management University, supervised by Prof. LIAO Lizi. His research focuses on conversational understanding, LLM safety, and multi-agent systems.
More About The Series
AI Wednesdays is Lorong AI’s weekly gathering, bringing together practitioners, researchers and innovators for technical discussions on research insights, product development and engineering practices.
Get involved: Learn more about Lorong AI | Speaker Sign-up | WhatsApp Community | LinkedIn | X