Sotabase

Career

· Growth GTM, Vert2026–
· Investment Banking Summer Analyst, J.P. Morgan2025–
· PhD Student, UC Berkeley2016–2021
· Research Assistant, UC Berkeley Robot Learning Lab2016–2021
· Research Assistant, UC Berkeley2014–
· Software Engineer, Apple
· Senior Software Engineer, Baidu, Inc.
· Master of Science, Beijing Forestry University
· J.P. Morgan
· Research Intern, OpenAI

Publications (57)

GPT-4 Technical Report
2023
21,596
cited
GPT-4o System Card
arXiv.org · 2024
2,980
cited
WILDS: A Benchmark of in-the-Wild Distribution Shifts
International Conference on Machine Learning · 2020
1,674
cited
When to Trust Your Model: Model-Based Policy Optimization
Neural Information Processing Systems · 2019
1,109
cited
MEMO: Test Time Robustness via Adaptation and Augmentation
Neural Information Processing Systems · 2021
453
cited
Adaptive Risk Minimization: Learning to Adapt to Domain Shift
Neural Information Processing Systems · 2020
246
cited
SOLAR: Deep Structured Representations for Model-Based Reinforcement Learning
International Conference on Machine Learning · 2018
229
cited
AVID: Learning Multi-Stage Tasks via Pixel-Level Translation of Human Videos
Robotics: Science and Systems · 2019
173
cited
Combining Model-Based and Model-Free Updates for Trajectory-Centric Reinforcement Learning
International Conference on Machine Learning · 2017
163
cited
Deep reinforcement learning for tensegrity robot locomotion
IEEE International Conference on Robotics and Automation · 2016
102
cited
Learning deep neural network policies with continuous memory states
IEEE International Conference on Robotics and Automation · 2015
95
cited
Adaptive Risk Minimization: A Meta-Learning Approach for Tackling Group Shift
arXiv.org · 2020
86
cited
OpenAI GPT-5 System Card
2025
46
cited
Towards Implicit Bias Detection and Mitigation in Multi-Agent LLM Interactions
Conference on Empirical Methods in Natural Language Processing · 2024
43
cited
SOLAR: Deep Structured Latent Representations for Model-Based Reinforcement Learning
arXiv.org · 2018
42
cited
Policy Learning with Continuous Memory States for Partially Observed Robotic Control
arXiv.org · 2015
19
cited
Losing Visual Needles in Image Haystacks: Vision Language Models are Easily Distracted in Short and Long Contexts
Conference on Empirical Methods in Natural Language Processing · 2024
9
cited
Out-of-Distribution Detection Methods Answer the Wrong Questions
arXiv.org · 2025
9
cited
When Every Token Counts: Optimal Segmentation for Low-Resource Language Models
COLING Workshops · 2024
6
cited
Seed-Free Synthetic Data Generation Framework for Instruction-Tuning LLMs: A Case Study in Thai
Annual Meeting of the Association for Computational Linguistics · 2024
5
cited
Sotabase
Marvin Zhang | Researcher Profile | Sotabase | Sotabase