Picture for Thomas L. Griffiths

Thomas L. Griffiths

VideoGameBench: Can Vision-Language Models complete popular video games?

Add code
May 23, 2025
Viaarxiv icon

Are Large Language Models Reliable AI Scientists? Assessing Reverse-Engineering of Black-Box Systems

Add code
May 23, 2025
Viaarxiv icon

Partner Modelling Emerges in Recurrent Agents (But Only When It Matters)

Add code
May 22, 2025
Viaarxiv icon

Using Reinforcement Learning to Train Large Language Models to Explain Human Decisions

Add code
May 16, 2025
Viaarxiv icon

Steering Risk Preferences in Large Language Models by Aligning Behavioral and Neural Representations

Add code
May 16, 2025
Viaarxiv icon

Predictability Shapes Adaptation: An Evolutionary Perspective on Modes of Learning in Transformers

Add code
May 14, 2025
Viaarxiv icon

Recovering Event Probabilities from Large Language Model Embeddings via Axiomatic Constraints

Add code
May 10, 2025
Viaarxiv icon

Toward Efficient Exploration by Large Language Model Agents

Add code
Apr 29, 2025
Viaarxiv icon

Identifying and Mitigating the Influence of the Prior Distribution in Large Language Models

Add code
Apr 17, 2025
Viaarxiv icon

Localized Cultural Knowledge is Conserved and Controllable in Large Language Models

Add code
Apr 14, 2025
Viaarxiv icon