Thousands of papers are published every day.
You shouldn't have to search for the ones you need.
Sarah Chen
University of Oxford, Department of Experimental Psychology
Research summary
Cognitive scientist studying judgment and decision-making, with a focus on how AI systems and large language models reason under uncertainty.
Interests
Active projects
llm-reasoning
How large language models reason under uncertainty
decision-making
Anchoring and bias in expert judgment
Key publications
Following
R. Williams, T. Nakamura, M. Hernandez, and 8 others
Sarah's sources
arXiv
Preprints
Journals
Causal Inference in Large Language Models: A Structural Approach
Zhang, L., Hernandez, M., & Park, S. · arXiv
Dimension scores
Sarah's Daily Papers · Feb 14, 2026
1,247 papers scanned · 12 selected · 3 projects matched
Causal Inference in Large Language Models: A Structural Approach
Zhang, L., Hernandez, M., & Park, S. · arXiv
Proposes a structural causal framework for evaluating reasoning in language models, with experiments on three benchmark datasets.
Pre-registered Replication of Anchoring Effects in Expert Judgment
Williams, R. & Okafor, A. · Psychological Science
A large-sample pre-registered replication that largely confirms anchoring effects among domain experts.
Network Dynamics of Scientific Collaboration Across Disciplines
Chen, W., Dubois, P., & Nakamura, T. · Nature Human Behaviour
Maps collaboration networks across 15 disciplines, finding that interdisciplinary bridges predict long-term citation impact.
+ 9 more papers in full digest
Your Library
47 papersCausal Inference in Large Language Models: A Structural Approach
Zhang, L., Hernandez, M., & Park, S. · arXiv · 2026
Pre-registered Replication of Anchoring Effects in Expert Judgment
Williams, R. & Okafor, A. · Psychological Science · 2026
Network Dynamics of Scientific Collaboration Across Disciplines
Chen, W., Dubois, P., & Nakamura, T. · Nature Human Behaviour · 2025
Bayesian Optimization for Experimental Design in Cognitive Science
Patel, A. & Kim, J. · Cognitive Science · 2026