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 […]

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 […]

Standard English variants exist as a way for all English speakers to communicate in a mutually intelligible fashion, and these standard language forms are generally associated with professionalism, official establishments, and authority. In fact, most media, communication, or services intended for the general public would use standardized English variants, which is why large language models […]

I’d like to begin by sharing some thoughts on Petter’s article concerning ChatGPT’s role in annotating political tweets, drawing parallels with my own experiences. This summer, one of my internship projects involved categorizing various country reports using a predefined set of categories, which numbered over twenty. My initial approach was to employ ChatGPT for zero-shot […]

Reading Deshpande et al.’s paper on toxicity in ChatGPT has only resulted in more questions about how ChatGPT comes to these responses when being assigned certain personas. Some results are unsurprising, such as having ChatGPT take the persona of Adolf Hitler, who people almost universally agree was a bad person and therefore, ChatGPT’s responses while […]

Last week’s debate concerned the prompt “AI Will Bring More Creativity and Innovation than Risks and Dangers”, with Colby, Fady, and Risabh affirming the claim and Louisa, Sophie, Seth and me opposing it. From the beginning, I was surprised by how opinions varied on specific topics related to AI, even within the teams. Both sides […]

At the end of Week 5, we had a debate on whether ChatGPT and similar AI tools would have net positive or negative impacts on higher education. Half of the class participated — I myself was a speaker on the negative side. The prompt recommended that we focus on the ideas put forth by sociologist […]