Recent advancements in large language models (LLMs) have paved the way for AI educational assistants in academic settings. However, AI assistants often respond differently than TAs, providing extensive explanations that may overwhelm students or inadvertently reveal more than intended. This study identifies the main differences between TA and LLM responses to students by using a four-class utterance classification system to compare the utterance distributions found in TA replies and in responses generated by Edison, a state-of-the-art AI educational assistant. Using this classification, striking distributional differences are observed: Edison produces far more Advance utterances, whereas TAs use many more React and Social Convention utterances. This research examines how differences in these distributions relate to response quality in student–TA interactions. Through prompt engineering, we align Edison's utterance distribution with TA patterns, producing responses that are more concise, directly address student questions, and avoid unnecessary elaboration.

In HCI, adult concerns about technologies for children have been studied extensively. However, less is known about what children themselves find concerning in everyday technologies. We examine children's technology-related fears by probing their use of the colloquial term "creepy." To understand children's perceptions of "creepy technologies," we conducted four participatory design sessions with children (ages 7 - 11) to design and evaluate creepy technologies, followed by interviews with the same children. We found that children's fear reactions emphasized physical harm and threats to their relationships (particularly with attachment figures). The creepy signals from technology the children described include: deception, lack of control, mimicry, ominous physical appearance, and unpredictability. Children acknowledged trusted adults will mediate the relationship between creepy technology signals and fear responses. Our work contributes a close examination of what children mean when they say a technology is "creepy." By treating these concerns as principal design considerations, developers can build systems that are more transparent about the risks they produce and more sensitive to the fears they may unintentionally raise.

We describe an adaptive method for extracting records from web pages. Our algorithm combines a weighted tree matching metric with clustering for obtaining data extraction patterns.We compare our method experimentally to the state-of-the-art, and show that our approach is very competitive for rigidly-structured records (such as product descriptions) and far superior for loosely-structured records (such as entrieson blogs).