Search for: All records

An advantage of Large Language Models (LLMs) is their contextualization capability – providing different responses based on student inputs like solution strategy or prior discussion, to potentially better engage students than standard feedback. We present a design and evaluation of a proof-of-concept LLM application to offer students dynamic and contextualized feedback. Specifically, we augment an Online Programming Exercise bot for a college-level Cloud Computing course with ChatGPT, which offers students contextualized reflection triggers during a collaborative query optimization task in database design. We demonstrate that LLMs can be used to generate highly situated reflection triggers that incorporate details of the collaborative discussion happening in context. We discuss in depth the exploration of the design space of the triggers and their correspondence with the learning objectives as well as the impact on student learning in a pilot study with 34 students. 

Fanfiction presents an opportunity as a data source for research in NLP, education, and social science. However, answering specific research questions with this data is difficult, since fanfiction contains more diverse writing styles than formal fiction. We present a text processing pipeline for fanfiction, with a fo- cus on identifying text associated with characters. The pipeline includes modules for character identification and coreference, as well as the attribution of quotes and narration to those characters. Additionally, the pipeline contains a novel approach to character coreference that uses knowledge from quote attribution to resolve pronouns within quotes. For each module, we evaluate the effectiveness of various approaches on 10 annotated fanfiction stories. This pipeline outperforms tools developed for formal fiction on the tasks of character coreference and quote attribution. 

In this paper, we present a co-design study with teachers to contribute towards the development of a technology-enhanced Artificial Intelligence (AI) curriculum, focusing on modeling unstructured data. We created an initial design of a learning activity prototype and explored ways to incorporate the design into high school classes. Specifically, teachers explored text classification models with the prototype and reflected on the exploration as a user, learner, and teacher. They provided insights about learning opportunities in the activity and feedback for integrating it into their teaching. Findings from qualitative analysis demonstrate that exploring text classification models provided an accessible and comprehensive approach for integrated learning of mathematics, language arts, and computing with the potential of supporting the understanding of core AI concepts including identifying structure within unstructured data and reasoning about the roles of human insight in developing AI technologies. 

The paper describes the first two field studies on using Rebo (Junior - non-adaptive) , and essentially shows that the reflection dialogue structure can lead to coherent and reflective questions, and does so in a majority of cases; and that contrary to expectations, engagement doesn't drop (engagement in the sense of coherence and reflectivity of conversations). 

In this paper, we present a co-design study with teachers to contribute towards development of a technology-enhanced Artificial Intelligence (AI) curriculum, focusing on modeling unstructured data. We created an initial design of a learning activity prototype and explored ways to incorporate the design into high school classes. Specifically, teachers explored text classification models with the prototype and reflected on the exploration as a user, learner, and teacher. They provided insights about learning opportunities in the activity and feedback for integrating it into their teaching. Findings from qualitative analysis demonstrate that exploring text classification models provided an accessible and comprehensive approach for integrated learning of mathematics, language arts, and computing with the potential of supporting the understanding of core AI concepts including identifying structure within unstructured data and reasoning about the roles of human insight in developing AI technologies.