Introduction This is my individual report on the group project that I did with Louisa and Dean. The original presentation can be found here. Our project attempted to evaluate whether LLMs could adequately recognize and simulate cognitive biases. We specifically tested the models on hostile attribution bias (HAB), but we hoped that our findings […]
Continue reading →The excerpt of Daston’s Rules was illuminating for me, as it articulated a clear approach towards the relationship and nuance between human and machine intelligence. I was particularly touched by the following two lines: “The inference drawn from the capacity of machines to calculate was not that machines were intelligent but rather that at least some intelligence […]
Continue reading →Debate Topic: Will AI bring more creativity and innovation or risks and dangers? Reflecting on the debate, it occurs to me that there were two different forms of creativity and innovation being referenced interchangeably that day, namely, human creativity and innovation on the one hand, and AI creativity and innovation on the other. The effects […]
Continue reading →For our group project, we undertook a detailed evaluation of GPT-4’s performance on American Mathematics Competition 12 (AMC 12) problems, utilizing two distinct prompting techniques: persona-based prompting and chain-of-thought prompting. For the persona-based approach, we initiated the conversation with the following system prompt: “You are a PhD graduate in Applied Mathematics from the University of […]
Continue reading →COGS 20100 Final Project Report (Fall 2023) Section 1: An Overview of Human Cognitive Biases Cognitive biases are systematic patterns of deviation from norm or rationality in judgment, where inferences about other people and situations may be illogical (Tversky & Kahneman, 1974). Some of the most prominent cognitive biases include (1) confirmation bias, where individuals […]
Continue reading →Project Report: Multimodal Evaluation of LLM’s Standardized Test Performance For the final project for our Perspectives on LLMs class, Fady Adal, Lihao Sun, Yushu Qiu and I (Jiaming Feng) worked together to design and evaluate GPT-4’s capability to answer standardized test questions that involve multimodal capabilities. Fady did background research on the topic; Lihao built […]
Continue reading →Title: Can large language models understand and evaluate hostile attribution biases? Introduction In history, machines have been built to be cold, unbiased, and objective. Cognitive biases, a systematic way of distorting one’s perceptions and judgments that give rise to individual’s subjective experiences, was thought to be an attribute unique to human beings. Among all the […]
Continue reading →Pre-preprint is available at: https://drive.google.com/file/d/1UYbU_txAfmuvWOyHUT2rWG9fNyzdOZtd/view?usp=sharing 1 Introduction The increasing prevalence of Large Language Models (LLMs) and Artificial Intelligence (AI) in contemporary society has led to the emergence of a wide array of evaluation benchmarks. Despite numerous efforts, we are faced with ever-expanding datasets encompassing a diverse range of tasks. The prevailing assumption that an increase […]
Continue reading →In the past couple of years, AI has been thrust into the spotlight especially with the release of ChatGPT in 2022. Some educators, professionals, and corporations are increasingly interested in using the technology to their advantage, while others are raising concerns over the technology’s limitations and dangers. Related to this, one question that our group […]
Continue reading →For a version with graphs, you can use this link: https://publuu.com/flip-book/330094/760112 Project Introduction: Our group’s final project focused on providing a multimodal evaluation of Large Language Models’ (LLM) cognitive capabilities, specifically their ability to understand and analyze graphs. Given time constraints, we primarily evaluated GPT-4, as it is currently one of the most popular LLM […]
Continue reading →