Sotabase

Career

· AI Institute Fellow, Schmidt Sciences2025–
· Postdoctoral Researcher, Stanford University2025–
· Resident, Anthropic2024–2025
· Research Intern, Allen Institute for AI2023–
· Research Intern, Google Research2022–
· Research Intern, Meta FAIR2021–
· PhD in Computer Science, The University of North Carolina at Chapel Hill2019–2024
· BS in Statistical Science, Duke University2015–2019

Publications (54)

Open Problems and Fundamental Limitations of Reinforcement Learning from Human Feedback
Trans. Mach. Learn. Res. · 2023
738
cited
Evaluating Explainable AI: Which Algorithmic Explanations Help Users Predict Model Behavior?
Annual Meeting of the Association for Computational Linguistics · 2020
334
cited
Does Localization Inform Editing? Surprising Differences in Causality-Based Localization vs. Knowledge Editing in Language Models
Neural Information Processing Systems · 2023
235
cited
Rethinking machine unlearning for large language models
Nature Machine Intelligence · 2024
219
cited
Foundational Challenges in Assuring Alignment and Safety of Large Language Models
arXiv.org · 2024
201
cited
Reasoning Models Don't Always Say What They Think
arXiv.org · 2025
195
cited
Can Sensitive Information Be Deleted From LLMs? Objectives for Defending Against Extraction Attacks
International Conference on Learning Representations · 2023
152
cited
GrIPS: Gradient-free, Edit-based Instruction Search for Prompting Large Language Models
Conference of the European Chapter of the Association for Computational Linguistics · 2022
150
cited
FastIF: Scalable Influence Functions for Efficient Model Interpretation and Debugging
Conference on Empirical Methods in Natural Language Processing · 2020
134
cited
Interpretable Image Recognition with Hierarchical Prototypes
AAAI Conference on Human Computation & Crowdsourcing · 2019
129
cited
Leakage-Adjusted Simulatability: Can Models Generate Non-Trivial Explanations of Their Behavior in Natural Language?
Findings · 2020
112
cited
The Out-of-Distribution Problem in Explainability and Search Methods for Feature Importance Explanations
Neural Information Processing Systems · 2021
105
cited
When Can Models Learn From Explanations? A Formal Framework for Understanding the Roles of Explanation Data
LNLS · 2021
91
cited
Do Language Models Have Beliefs? Methods for Detecting, Updating, and Visualizing Model Beliefs
arXiv.org · 2021
89
cited
Methods for Measuring, Updating, and Visualizing Factual Beliefs in Language Models
Conference of the European Chapter of the Association for Computational Linguistics · 2023
58
cited
Can Language Models Teach Weaker Agents? Teacher Explanations Improve Students via Theory of Mind
arXiv.org · 2023
26
cited
Summarization Programs: Interpretable Abstractive Summarization with Neural Modular Trees
International Conference on Learning Representations · 2022
20
cited
Are Hard Examples also Harder to Explain? A Study with Human and Model-Generated Explanations
Conference on Empirical Methods in Natural Language Processing · 2022
18
cited
VisFIS: Visual Feature Importance Supervision with Right-for-the-Right-Reason Objectives
Neural Information Processing Systems · 2022
15
cited
Low-Cost Algorithmic Recourse for Users With Uncertain Cost Functions
arXiv.org · 2021
12
cited
Sotabase
Peter Hase | Researcher Profile | Sotabase | Sotabase