1. Tango 2: Aligning Diffusion-based Text-to-Audio Generations through Direct Preference Optimization, 2024
2. Inference Time Alignment with Reward-Guided Tree Search, 2024
3. TangoFlux: Super Fast and Faithful Text to Audio Generation with Flow Matching and Clap-Ranked Preference Optimization, 2024
4. NORA: A Small Open-Sourced Generalist Vision Language Action Model for Embodied Tasks, 2025
Awards & Projects:
1. Darwin accepted to NAACL 2025 (Oral)
2. Tango2 accepted to ACM MM 2024 (Oral)
3. Released projects like Nora, TangoFlux, etc.
Research Experience
Conducted research under Prof. Roy Ka-Wei Lee and Prof. Soujanya Poria during undergraduate period. Current work: Working on reinforcement learning with action conditioned world models.
Education
Bachelor's Degree: Singapore University of Technology and Design (SUTD), Major: Computer Science, Time: May 2024; PhD Student: Nanyang Technological University (NTU), Advisor: Prof. Soujanya Poria, Time: Present.
Background
Research Interest: MultiModals and Vision Language Action Models. Brief Introduction: Earned Bachelor’s degree in Computer Science from SUTD, currently a PhD student at NTU.
Miscellany
Personal Interests: Feel free to email me to chat about reinforcement learning research.