Sotabase

Career

· Research Scientist, FAR.AI2024–
· Research Intern, DeepMind2022–2022
· PhD - Computer Science, Stanford University2018–2024
· Visiting Scholar, Centre for Human-Compatible AI, University of California, Berkeley2017–2017
· Visiting Scholar, Future of Humanity Institute, University of Oxford2017–2018
· MEng - Computer Science, University of Cambridge2016–2017
· BA - Natural Sciences (Physics), University of Cambridge2013–2016
· FAR AI

Publications (0)

IQ-Learn: Inverse soft-Q Learning for Imitation
Neural Information Processing Systems · 2021
230
cited
On the Opportunities and Challenges of Foundation Models for Geospatial Artificial Intelligence
arXiv.org · 2023
154
cited
Parallelizing Linear Recurrent Neural Nets Over Sequence Length
International Conference on Learning Representations · 2017
153
cited
Geo-knowledge-guided GPT models improve the extraction of location descriptions from disaster-related social media messages
International Journal of Geographical Information Science · 2023
138
cited
BCD Nets: Scalable Variational Approaches for Bayesian Causal Discovery
Neural Information Processing Systems · 2021
84
cited
Towards a foundation model for geospatial artificial intelligence (vision paper)
SIGSPATIAL/GIS · 2022
81
cited
LMPriors: Pre-Trained Language Models as Task-Specific Priors
arXiv.org · 2022
62
cited
KGGen: Extracting Knowledge Graphs from Plain Text with Language Models
arXiv.org · 2025
26
cited
Simulation of Plants in Buildings; Incorporating Plant-Air Interactions in Building Energy Simulation
Building Simulation Conference Proceedings · 2015
16
cited
SequenceMatch: Imitation Learning for Autoregressive Sequence Modelling with Backtracking
International Conference on Learning Representations · 2023
15
cited
Evaluation of Cardiovascular Parameters of a Selenium‐Based Antihypertensive Using Pulsed Doppler Ultrasound
Journal of Cardiovascular Pharmacology · 2001
11
cited
FHI Oxford Technical Report # 2018-2 Predicting Human Deliberative Judgments with Machine Learning
2018
9
cited
Planning in a recurrent neural network that plays Sokoban
2024
9
cited
A physics-informed machine learning model for the prediction of drop breakup in two-phase flows
International Journal of Multiphase Flow · 2024
6
cited
Exploring Hierarchy-Aware Inverse Reinforcement Learning
arXiv.org · 2018
5
cited
No, of course I can! Refusal Mechanisms Can Be Exploited Using Harmless Fine-Tuning Data
2025
5
cited
Preference Learning with Lie Detectors can Induce Honesty or Evasion
arXiv.org · 2025
4
cited
Privacy-Constrained Policies via Mutual Information Regularized Policy Gradients
International Conference on Artificial Intelligence and Statistics · 2020
4
cited
AI Companies Should Report Pre- and Post-Mitigation Safety Evaluations
arXiv.org · 2025
3
cited
Beyond Bayes-optimality: meta-learning what you know you don't know
arXiv.org · 2022
2
cited
Sotabase