Discrete Flow Matching Policy Optimization
Introduces DoMinO, an RL framework that treats discrete flow-matching inference as an inner MDP. Outperforms prior state-of-the-art by 10.6% on regulatory DNA design and 15.4% on …
m.-su
Po-Chung (Bryan) Hsieh
M.S. Student in Electrical and Computer Engineering
I am an M.S. student in Electrical and Computer Engineering at UC San Diego, with a B.S. in Electrical Engineering from National Taiwan University. I build AI agents and reinforcement learning methods for scientific discovery, including multi-agent systems for biomedical feature discovery and policy optimization for discrete generative models, and I previously worked as an LLM engineer intern at the Inventec Artificial Intelligence Center.
•
•
1 min read
The Binding Effect: Analyzing How Multi-Dimensional Cues Form Gender Bias in Instruction TTS
Accepted to INTERSPEECH 2026.
k.-y.-chen
Dual-Process Scaffold Reasoning for Enhancing LLM Code Debugging
A dual-process scaffold reasoning framework for LLM code debugging, reaching an 88.91% pass rate on DebugBench.
