Publications
2026
-
Monitoring and Discovering Reward Hacking with Internal Representations during LLM EvaluationsarXiv preprint arXiv:2609.19101, 2026. -
AUTO-PSYCH: Automating the science of mind using agent-driven theory discovery and experimentationCOLM Workshop on Language Models for Scientific Discovery, 2026. (Outstanding Paper Award) -
CORE: Contrastive Reflection Enables Rapid Improvements in ReasoningarXiv preprint arXiv:2605.28742, 2026. -
Leveraging Speech to Identify Signatures of Insight and Transfer in Problem SolvingCogsci, 2026.
2025
2024
-
Rethinking Out-of-Distribution Detection for Reinforcement Learning: Advancing Methods for Evaluation and DetectionAAMAS, 2024. (Oral Presentation) -
Using adaptive intrinsic motivation in RL to model learning across developmentNeurIPS Intrinsically Motivated Open-ended Learning Workshop, 2024.
