Cognitive Bias
Cognitive biases are systematic errors in thinking that affect decision-making, and recent research focuses on identifying and mitigating these biases in large language models (LLMs). Studies employ various methods, including multi-agent systems, prompt engineering, and Bayesian opinion aggregation, to analyze how LLMs exhibit biases like anchoring, framing, and loss aversion, mirroring human cognitive processes. Understanding and addressing these biases is crucial for improving the reliability and fairness of LLMs across diverse applications, from news recommendation to high-stakes decision support in fields like healthcare.
Papers
Words That Stick: Predicting Decision Making and Synonym Engagement Using Cognitive Biases and Computational Linguistics
Nimrod Dvir, Elaine Friedman, Suraj Commuri, Fan Yang, Jennifer Romano
A Predictive Model of Digital Information Engagement: Forecasting User Engagement With English Words by Incorporating Cognitive Biases, Computational Linguistics and Natural Language Processing
Nimrod Dvir, Elaine Friedman, Suraj Commuri, Fan yang, Jennifer Romano