Chenglong Wang
Post-graduate student @ Northeastern University
reinforcement learning
knowledge distillation
glue
sequence generation
large language model
actor-critic
dynamic sampling
instruction tuning
evol instruct
5
presentations
1
number of views
SHORT BIO
I am currently a third-year Ph.D. student at the Natural Language Processing Laboratory, Department of Computer Science and Technology, Northeastern University, China, under the guidance of Professor Tong Xiao and Professor Jingbo Zhu.
I completed my undergraduate studies at Northeastern University and obtained my bachelor's degree in Computer Science and Technology. Subsequently, I pursued my master's degree, also at Northeastern University, graduating in 2020.
My research interests focus on modeling methods for deep reinforcement learning on sequence generation tasks, with a particular emphasis on two key areas: 1) enhancing the training efficiency of sequence generation models based on reinforcement learning, and 2) constructing and training reward models for deep reinforcement learning.
Presentations
RoVRM: A Robust Visual Reward Model Optimized via Auxiliary Textual Preference Data
Yang Gan and 9 other authors
Revealing the Parallel Multilingual Learning within Large Language Models
Yongyu Mu and 10 other authors
The NiuTrans System for the WMT 2021 Efficiency Task
Chenglong Wang
Improved Knowledge Distillation for Pre-trained Language Models via Knowledge Selection
Chenglong Wang
ESRL: Efficient Sampling-Based Reinforcement Learning for Sequence Generation
Chenglong Wang and 7 other authors