Zhang-Wei Hong

PhD student, Massachusetts Institute of Technology

Joined

December 2017

Names

Zhang-Wei Hong (Preferred)

Emails

****@gapp.nthu.edu.tw (Confirmed)

****@mit.edu (Confirmed)

Personal Links

Google Scholar

DBLP

Career & Education History

PhD student

Massachusetts Institute of Technology (mit.edu)

2020 – Present

Internship

Research, Microsoft (research.microsoft.com)

2022 – 2022

Internship

Preferred Networks, Inc. (preferred.jp)

2019 – 2019

MS student

National Tsing Hua University (nthu.edu.tw)

2017 – 2018

Advisors, Relations & Conflicts

PhD Advisor

Pulkit Agrawal

2020 – 2024

Expertise

Reinforcement learning

2016 – Present

Publications

Rethinking Exploration Through Context for RL
Sunshine Jiang, Zhang-Wei Hong, John Marangola, Nitish Dashora, Richard Li, Pulkit Agrawal
- ICRA 2026: From Data to Decisions Poster
- Readers: Everyone
Pushing Forward Pareto Frontiers of Proactive Agents with Behavioral Agentic Optimization
Yihang Yao, Zhepeng Cen, Haohong Lin, Shiqi Liu, Zuxin Liu, Jiacheng Zhu, Zhang-Wei Hong, Laixi Shi, Ding Zhao
- LLA 2026 Poster
- Readers: Everyone
Tailored Primitive Initialization is the Secret Key to Reinforcement Learning
Yihang Yao, Guangtao Zeng, Raina Wu, Yang Zhang, Ding Zhao, Zhang-Wei Hong, Chuang Gan
- ICLR 2026 Conference Withdrawn Submission
- Readers: Everyone
BOAD: Discovering Hierarchical Software Engineering Agents via Bandit Optimization
Iris Xu, Guangtao Zeng, Zexue He, Charles Jin, Aldo Pareja, Dan Gutfreund, Chuang Gan, Zhang-Wei Hong
- ICLR 2026 Poster
- Readers: Everyone
GPTOpt: Towards Efficient LLM-based Black-Box Optimization
Jamison Meindl, Yunsheng Tian, Tony Cui, Veronika Thost, Zhang-Wei Hong, Jie Chen, Wojciech Matusik, Mina Konakovic Lukovic
- Submitted to ICLR 2026
- Readers: Everyone
EvoScale: Evolutionary Test-Time Scaling for Software Engineering
Guangtao Zeng, Maohao Shen, Delin Chen, Zhenting Qi, Subhro Das, Dan Gutfreund, David Daniel Cox, Gregory W. Wornell, Wei Lu, Zhang-Wei Hong, Chuang Gan
- Submitted to ICLR 2026
- Readers: Everyone
A Reward-Free Viewpoint on Multi-Objective Reinforcement Learning
Ying-Tu Chen, Wei Hung, Bing-Shu Wu, Zhang-Wei Hong, Ping-Chun Hsieh
- ICLR 2026 Poster
- Readers: Everyone
Composition-Grounded Data Synthesis for Visual Reasoning
Xinyi Gu, Jiayuan Mao, Zhang-Wei Hong, Zhuoran Yu, Pengyuan Li, Dhiraj Joshi, Rogerio Feris, Zexue He
- ICLR 2026 Poster
- Readers: Everyone
Large Pretraining Datasets Don't Guarantee Robustness after Fine-Tuning
Jaedong Hwang, Brian Cheung, Zhang-Wei Hong, Akhilan Boopathy, Pulkit Agrawal, Ila R Fiete
- Submitted to ICLR 2026
- Readers: Everyone
Towards Generating Stable Materials via Large Language Models with Reinforcement Learning Finetuning
Zhang-Wei Hong, Nofit Segal, Aviv Netanyahu, Hoje Chun, Rafael Gomez-Bombarelli, Pulkit Agrawal
- NeurIPS2025-AI4Science Poster
- Readers: Everyone

View all 69 publications

Co-Authors

View all 132 co-authors