Selected publications include 'GThinker: Towards General Multimodal Reasoning via Cue-Guided Rethinking', 'R1-ShareVL: Incentivizing Reasoning Capability of Multimodal Large Language Models via Share-GRPO'. Involved in projects such as Valley2, 'Decoder-Only LLMs can be Masked Auto-Encoders' etc.
Research Experience
Staff Algorithm Expert at ByteDance, working on LLM for E-commerce, specifically the Valley series. Senior Algorithm Expert at Alibaba Group, developed PAI-EasyTransfer framework and EasyNLP toolkit. Intern at Google Inc., worked on Random Features for Large-Scale Kernel Machines. Visiting Scholar at Carnegie Mellon University, focused on Natural Language Processing and Machine Learning.
Education
PhD: School of Information Systems, Singapore Management University, supervised by Associate Prof. Jing Jiang and Prof Lim Ee-peng; PhD Overseas Training Residency: Language Technologies Institute, Carnegie Mellon University, supervised by Noah Smith and Alex Smola; Postdoc: Zhejiang University & Alibaba Group, advisors: Cai Deng (Zhejiang University), Jingren Zhou (Alibaba Group).
Background
Currently a Tech leader at Bytedance, focusing on LLM and MLLM for E-commerce. Previously, a senior algorithm expert at Alibaba Cloud, responsible for building NLP and Transfer learning toolkits, namely EasyNLP and EasyTransfer. Research interests include Text Mining and Analytics, Deep Learning, Reinforcement Learning, and Transfer Learning.
Miscellany
Active on Zhihu platform, personal website: https://minghui.github.io/.