Shambhavi Bhushan

Tutoring improves student achievement, but identifying and studying what tutoring actions are most associated with student learning at scale based on audio transcriptions is an open research problem. This present study investigates the feasibility and scalability of using generative AI to identify and evaluate specific tutor moves in real-life math tutoring. We analyze 50 randomly selected transcripts of college-student remote tutors assisting middle school students in mathematics. Using GPT-4, GPT-4o, GPT-4-turbo, Gemini-1.5-pro, and LearnLM, we assess tutors' application of two tutor skills: delivering effective praise and responding to student math errors. All models reliably detected relevant situations, for example, tutors providing praise to students (94-98% accuracy) and a student making a math error (82-88% accuracy) and effectively evaluated the tutors' adherence to tutoring best practices, aligning closely with human judgments (83-89% and 73-77%, respectively). We propose a cost-effective prompting strategy and discuss practical implications for using large language models to support scalable assessment in authentic settings. This work further contributes LLM prompts to support reproducibility and research in AI-supported learning.

The increasing availability of large language models (LLMs) has raised concerns about their potential misuse in online learning. While tools for detecting LLM-generated text exist and are widely used by researchers and educators, their reliability varies. Few studies have compared the accuracy of detection methods, defined criteria to identify content generated by LLM, or evaluated the effect on learner performance from LLM misuse within learning. In this study, we define LLM-generated text within open responses as those produced by any LLM without paraphrasing or refinement, as evaluated by human coders. We then fine-tune GPT-4o to detect LLM-generated responses and assess the impact on learning from LLM misuse. We find that our fine-tuned LLM outperforms the existing AI detection tool GPTZero, achieving an accuracy of 80% and an F1 score of 0.78, compared to GPTZero's accuracy of 70% and macro F1 score of 0.50, demonstrating superior performance in detecting LLM-generated responses. We also find that learners suspected of LLM misuse in the open response question were more than twice as likely to correctly answer the corresponding posttest MCQ, suggesting potential misuse across both question types and indicating a bypass of the learning process. We pave the way for future work by demonstrating a structured, code-based approach to improve LLM-generated response detection and propose using auxiliary statistical indicators such as unusually high assessment scores on related tasks, readability scores, and response duration. In support of open science, we contribute data and code to support the fine-tuning of similar models for similar use cases.

Assessing learners in ill-defined domains, such as scenario-based human tutoring training, is an area of limited research. Equity training requires a nuanced understanding of context, but do contemporary large language models (LLMs) have a knowledge base that can navigate these nuances? Legacy transformer models like BERT, in contrast, have less real-world knowledge but can be more easily fine-tuned than commercial LLMs. Here, we study whether fine-tuning BERT on human annotations outperforms state-of-the-art LLMs (GPT-4o and GPT-4-Turbo) with few-shot prompting and instruction. We evaluate performance on four prediction tasks involving generating and explaining open-ended responses in advocacy-focused training lessons in a higher education student population learning to become middle school tutors. Leveraging a dataset of 243 human-annotated open responses from tutor training lessons, we find that BERT demonstrates superior performance using an offline fine-tuning approach, which is more resource-efficient than commercial GPT models. We conclude that contemporary GPT models may not adequately capture nuanced response patterns, especially in complex tasks requiring explanation. This work advances the understanding of AI-driven learner evaluation under the lens of fine-tuning versus few-shot prompting on the nuanced task of equity training, contributing to more effective training solutions and assisting practitioners in choosing adequate assessment methods.

Addressing negative self-talk by students, such as responding to a student when saying, "I am dumb" or "I can't do this" can be difficult for even the most experienced tutor. Despite potential tutor learning from scenario-based lessons on this topic, human-graded assessment remains time-consuming. Leveraging generative AI for evaluating textual responses in online training presents a scalable solution. Research suggests a tutor validates student's feelings when they speak negatively of themselves, e.g., by a tutor responding, "I understand how you feel" or "I recognize this is difficult." This ongoing work assesses the performance of 60 undergraduate tutors within an online lesson on enhancing tutors' abilities to respond to students engaging in negative self-talk. We find statistically significant tutor learning gains from pretest to posttest. Additionally, we describe a method of using generative AI for assessing tutors' responses to predict the best approach and subsequently explain the rationale behind it. Using the large language model GPT-4, we find high absolute performance when evaluating tutor responses involving predicting (F1 = 0.85) and explaining (F1 = 0.83) the best approach. Minor improvements are needed to the lesson itself. A future goal of this work is to fully develop automated systems of assessing tutor learning attending to barriers to students' motivation and doing so at scale.