The paper delves into the intersection of Large Language Models (LLMs), Human-Computer Interaction (HCI), and Education by conducting surveys and interviews at undergraduate engineering universities in India. The study specifically focuses on the academic usage of ChatGPT, a popular LLM, among students and instructors. Through 1306 survey responses, 112 student interviews, and 27 instructor interviews, the research provides valuable insights into current usage patterns as well as perceived benefits, threats, and challenges of LLMs in educational settings. These findings offer recommendations for enhancing the adoption of LLMs among students and instructors. In the related work section, a literature review is presented on topics such as prompt engineering with LLMs, AI in Education, LLMs in Education, and Student and Instructor Perspectives on LLMs. This includes discussions on various studies that highlight the influence of prompt quality on LLM responses, adaptability of large language models for mobile interfaces, challenges faced by users in understanding model syntax and capabilities. Additionally, positive experiences with interactive applications like chatbots for public health interventions are also discussed. Furthermore, studies examining the impact of LLM-powered writing assistants on users' opinions and writings are explored. These studies emphasize the importance of carefully monitoring and engineering opinions embedded in LLMs to ensure their responsible use. Challenges identified include issues related to communication style preferences when using code assistants in computational notebooks. Overall,and beyond.
- - The paper explores the intersection of Large Language Models (LLMs), Human-Computer Interaction (HCI), and Education through surveys and interviews at undergraduate engineering universities in India.
- - Specifically focuses on the academic usage of ChatGPT among students and instructors.
- - Research includes 1306 survey responses, 112 student interviews, and 27 instructor interviews to provide insights into current usage patterns, benefits, threats, and challenges of LLMs in educational settings.
- - Findings offer recommendations for enhancing adoption of LLMs among students and instructors.
- - Literature review covers topics such as prompt engineering with LLMs, AI in Education, LLMs in Education, and Student and Instructor Perspectives on LLMs.
- - Discussions include studies on prompt quality's influence on LLM responses, adaptability of large language models for mobile interfaces, challenges faced by users in understanding model syntax and capabilities.
- - Positive experiences with interactive applications like chatbots for public health interventions are highlighted.
- - Studies examining the impact of LLM-powered writing assistants on users' opinions and writings are explored to emphasize responsible use.
SummaryThe paper looks at how big language models (LLMs), human-computer interaction (HCI), and education connect by asking questions and talking to people at engineering schools in India. They focus on how students and teachers use ChatGPT for school work. The research involved surveys from 1306 people, interviews with 112 students, and 27 teachers to learn about how LLMs are used in schools. The findings suggest ways to make LLMs more popular among students and teachers. They also talk about different topics like using prompts with LLMs, AI in education, and what students and teachers think about LLMs.
Definitions- Large Language Models (LLMs): Advanced computer programs that can understand and generate human-like text.
- Human-Computer Interaction (HCI): How people interact with computers or technology.
- Education: The process of learning in schools or universities.
- Surveys: Asking a group of people questions to gather information.
- Interviews: Talking to someone to ask them questions and get their thoughts or opinions.
Introduction:
The intersection of Large Language Models (LLMs), Human-Computer Interaction (HCI), and Education has become a topic of increasing interest in recent years. With the rise of artificial intelligence and natural language processing, LLMs have gained popularity for their ability to generate human-like text responses. This has led to their use in various fields, including education, where they are being explored as potential tools for enhancing learning experiences.
Research Paper Overview:
In this research paper, titled "Exploring the Usage Patterns and Perceptions of Large Language Models in Educational Settings", authors Anirudh Srinivasan, Prateek Jain, Shashank Singhania, and Ponnurangam Kumaraguru delve into the current usage patterns and perceptions of LLMs among students and instructors at undergraduate engineering universities in India. The study specifically focuses on ChatGPT, a popular LLM used for academic purposes.
Methodology:
To gather data for their research, the authors conducted surveys and interviews with 1306 students and 27 instructors from various engineering universities in India. In addition to this, they also conducted 112 student interviews to gain deeper insights into the usage patterns and perceptions of LLMs among users.
Findings:
The research paper presents valuable insights into the current usage patterns of LLMs among students and instructors. It highlights that while most students use LLMs primarily for generating text responses or completing assignments quickly, instructors tend to use them more for grading assignments or providing feedback on student work.
Perceived Benefits:
The study also sheds light on some perceived benefits of using LLMs in educational settings. These include saving time by automating tasks such as grading assignments or generating text responses quickly. Additionally, it is believed that using LLMs can improve writing skills by providing suggestions or corrections while typing.
Threats:
However, along with benefits come potential threats associated with using LLMs in education. The research paper identifies some of these threats, including the risk of plagiarism and over-reliance on LLMs for completing assignments or generating text responses.
Challenges:
The authors also highlight various challenges faced by users when using LLMs in educational settings. These include issues related to understanding model syntax and capabilities, as well as communication style preferences when using code assistants in computational notebooks.
Related Work:
In the related work section, the research paper provides a literature review on topics such as prompt engineering with LLMs, AI in Education, LLMs in Education, and Student and Instructor Perspectives on LLMs. This includes discussions on various studies that have been conducted to explore the impact of prompt quality on LLM responses, adaptability of large language models for mobile interfaces, and challenges faced by users in understanding model syntax and capabilities.
Additionally, the article also discusses positive experiences with interactive applications like chatbots for public health interventions. It highlights how these applications have successfully used LLMs to engage with users and provide them with relevant information.
Furthermore, studies examining the impact of LLM-powered writing assistants on users' opinions and writings are explored. These studies emphasize the importance of carefully monitoring and engineering opinions embedded in LLMs to ensure their responsible use.
Conclusion:
Overall, this research paper provides valuable insights into current usage patterns as well as perceived benefits, threats, and challenges associated with using Large Language Models (LLMs) in educational settings. The findings offer recommendations for enhancing the adoption of LLMs among students and instructors while emphasizing responsible use practices. With further advancements in technology and increased integration of AI tools like LLMs into education systems worldwide, it is crucial to continue exploring their potential benefits while addressing any potential risks they may pose.