Sotabase

Career

· PhD in Computer Science, UC Berkeley2020–

Publications (22)

Jailbroken: How Does LLM Safety Training Fail?
Neural Information Processing Systems · 2023
1,483
cited
Human-level play in the game of Diplomacy by combining language models with strategic reasoning
Science · 2022
476
cited
Optimal Robustness-Consistency Trade-offs for Learning-Augmented Online Algorithms
Neural Information Processing Systems · 2020
112
cited
More Than a Toy: Random Matrix Models Predict How Real-World Neural Representations Generalize
International Conference on Machine Learning · 2022
88
cited
Better and Simpler Learning-Augmented Online Caching
International Workshop and International Workshop on Approximation, Randomization, and Combinatorial Optimization. Algorithms and Techniques · 2020
73
cited
Covert Malicious Finetuning: Challenges in Safeguarding LLM Adaptation
International Conference on Machine Learning · 2024
64
cited
Predicting Out-of-Distribution Error with the Projection Norm
International Conference on Machine Learning · 2022
51
cited
Learning Equilibria in Matching Markets from Bandit Feedback
Neural Information Processing Systems · 2021
44
cited
Learning in Stackelberg Games with Non-myopic Agents
ACM Conference on Economics and Computation · 2022
40
cited
TCT: Convexifying Federated Learning using Bootstrapped Neural Tangent Kernels
Neural Information Processing Systems · 2022
33
cited
Temperature Self-Calibration of Always-On, Field-Deployed Ion-Selective Electrodes Based on Differential Voltage Measurement.
ACS Sensors · 2022
21
cited
Designing Approximately Optimal Search on Matching Platforms
ACM Conference on Economics and Computation · 2021
18
cited
Optimal Las Vegas Approximate Near Neighbors in 𝓁p
ACM-SIAM Symposium on Discrete Algorithms · 2018
12
cited
Allocation for Social Good: Auditing Mechanisms for Utility Maximization
ACM Conference on Economics and Computation · 2019
11
cited
X-Guard: Multilingual Guard Agent for Content Moderation
arXiv.org · 2025
11
cited
An Interscholastic Network To Generate LexA Enhancer Trap Lines in Drosophila
G3: Genes, Genomes, Genetics · 2019
9
cited
Varying the Number of Signals in Matching Markets
Workshop on Internet and Network Economics · 2018
8
cited
Learning Equilibria in Matching Markets with Bandit Feedback
Journal of the ACM · 2023
3
cited
Optimal Las Vegas Approximate Near Neighbors in l_p
ACM-SIAM Symposium on Discrete Algorithms · 2019
1
cited
OpenAI Research Engineer Interviews (Multiple Choice)
Sotabase
Alexander Wei | Researcher Profile | Sotabase | Sotabase