Research

At GLOW.AI, we are interested in learning and optimization on data-centric AI, collaborative AI, automated AI, and AI for science problems, their applications to LLMs & MLLMs, among others. See below for the research themes we work on.

Data-Centric AI

Data selection, data valuation, knowledge distillation, data attribution, machine unlearning, data ownership, domain generalization

Agents in ML

Federated learning, collaborative ML, incentives, reinforcement learning

Automated AI

Active learning, Bayesian & zeroth-order optimization, neural architecture search, meta-learning

AI for Science

Experimental design, physics-informed ML, quantum ML, precision agriculture

Recent News

See the list of our publications here.

Dec 2025 MEM1 is accepted for oral presentation (top 1%) at the NeurIPS 2025 Workshop on Multi-Turn Interactions in LLMs. Read more here.
Nov 2025 Invited to participate in the prestigious Dagstuhl Seminar 25451 on Bayesian Optimization. Read more here.
Jul 2025 Congrats! to Jingtan Wang, Zhiliang Chen, Gregory Lau, Xiaoqiang Lin, and Rui Qiao for receiving the Research Achievement Award. Read more here.
May 2025 Presented Data-Centric AI for Large Language Foundation Models as an invited speaker at the 16th Seoul Forum 2025 organised by the Seoul Economic Daily and the TTIC Summer Workshop on Incentives for Collaborative Learning and Data Sharing, Chicago, IL, Aug 13 - 15, 2025. Read more here.
Apr 2025 Presented Data-Centric AI Research @ GLOW.AI as an invited speaker in the ICLR 2025 Workshop on Navigating and Addressing Data Problems for Foundation Models (DATA-FM). Read more here.
Apr 2025 Presented Efficient Prompt Optimization for Adaptive LLMs: Instruction and Exemplar Tuning via Neural Bandits as an invited speaker in the ICLR 2025 Workshop on Scalable Optimization for Efficient and Adaptive Foundation Models (SCOPE). Read more here.
Apr 2025 Presented PIED: Physics-Informed Experimental Design for Inverse Problems as an invited speaker in the ICLR 2025 Workshop on XAI4Science: From Understanding Model Behavior to Discovering New Scientific Knowledge. Read more here.
Feb 2025 Our AI Singapore's National Multimodal LLM Programme research grant proposal (~S$2.9M for 3 years) on Towards Inverse Problems for Foundation Models is accepted! It is published as a position paper titled "Uncover Scaling Laws for Large Language Models via Inverse Problems" in EMNLP 2025. Read more here.
Feb 2025 I am a recipient of the Faculty Teaching Excellence Award 2025 in the School of Computing, NUS. Read more here.
Jan 2025 Congrats! to Sebastian Shenghong Tay for receiving the prestigious Dean's Graduate Research Excellence Award, and also to Zijian Zhou and Apivich Hemachandra for receiving the Research Achievement Award. Read more here.
Jul 2024 PINNACLE 🏔️ won the best paper award (out of 225 submissions) at the ICML 2024 Workshop on AI for Science! Read more here.
Jul 2024 Our AI Visiting Professorship Proposal (~S$4M for 3 years) on Data-Centric Machine Learning at Scale with Pang Wei Koh is accepted! It is featured on pages 29-30 in Issue 1 (Aug 2024) of Ignite, a new biannual magazine from the Office of the Deputy President (Research and Technology) of NUS. It is also published as a position paper titled "Data-Centric AI in the Age of Large Language Models" in EMNLP 2024. Read more here.
Nov 2025 Bryan served as a Senior Area Chair of ICML 2026, ICLR 2026, and an Area Chair of AAAI 2026, IJCAI 2025, AISTATS 2025, NeurIPS 2025.