Sotabase

Career

· Research Scientist, University of Oxford2024–

Publications (13)

Learn what matters: cross-domain imitation learning with task-relevant embeddings
Neural Information Processing Systems · 2022
27
cited
Extracting Reward Functions from Diffusion Models
Neural Information Processing Systems · 2023
19
cited
Rethinking Out-of-Distribution Detection for Reinforcement Learning: Advancing Methods for Evaluation and Detection
Adaptive Agents and Multi-Agent Systems · 2024
16
cited
Illusory Attacks: Information-theoretic detectability matters in adversarial attacks
International Conference on Learning Representations · 2022
12
cited
Learning Altruistic Behaviours in Reinforcement Learning without External Rewards
International Conference on Learning Representations · 2021
10
cited
Illusory Attacks: Detectability Matters in Adversarial Attacks on Sequential Decision-Makers
2023
6
cited
Reinforcement Learning for Quantum Control under Physical Constraints
International Conference on Machine Learning · 2025
4
cited
Select2Plan: Training-Free ICL-Based Planning Through VQA and Memory Retrieval
IEEE Robotics and Automation Letters · 2024
4
cited
Illusionary Attacks on Sequential Decision Makers and Countermeasures
arXiv.org · 2022
3
cited
Select to Perfect: Imitating desired behavior from large multi-agent data
International Conference on Learning Representations · 2024
3
cited
HelloFresh: LLM Evaluations on Streams of Real-World Human Editorial Actions across X Community Notes and Wikipedia edits
Annual Meeting of the Association for Computational Linguistics · 2024
2
cited
TuCo: Measuring the Contribution of Fine-Tuning to Individual Responses of LLMs
International Conference on Machine Learning · 2025
1
cited
High Accuracy, Less Talk (HALT): Reliable LLMs through Capability-Aligned Finetuning
arXiv.org · 2025
Sotabase
Tim Franzmeyer | Researcher Profile | Sotabase | Sotabase