Search for: All records

in mathematics education, and researchers often turn to advanced natural language processing (NLP) models to analyze classroom dialogues from multiple perspectives. However, utterance-level discourse analysis encounters two primary challenges: (1) multifunctionality, where a single utterance may serve multiple purposes that a single tag cannot capture, and (2) the exclusion of many utterances from domain-specific discourse move classifications, leading to their omission in feedback. To address these challenges, we proposed a multi-perspective discourse analysis that integrates domain-specific talk moves with dialogue act (using the flattened multi-functional SWBD-MASL schema with 43 tags) and discourse relation (applying Segmented Discourse Representation Theory with 16 relations). Our top-down analysis framework enables a comprehensive understanding of utterances that contain talk moves, as well as utterances that do not contain talk moves. This is applied to two mathematics education datasets: TalkMoves (teaching) and SAGA22 (tutoring). Through distributional unigram analysis, sequential talk move analysis, and multi-view deep dive, we discovered meaningful discourse patterns, and revealed the vital role of utterances without talk moves, demonstrating that these utterances, far from being mere fillers, serve crucial functions in guiding, acknowledging, and structuring classroom discourse. These insights underscore the importance of incorporating discourse relations and dialogue acts into AI-assisted education systems to enhance feedback and create more responsive learning environments. Our framework may prove helpful for providing human educator feedback, but also aiding in the development of AI agents that can effectively emulate the roles of both educators and students. 

This paper shares a synthesis of the literature related to the application of a relationships-first approach to high-dosage math tutoring. In the context of our research, high-dosage tutoring is delivered multiple times per week during the school day by paraprofessionals who work with students in historically under-resourced schools. We apply a critical perspective to frame the importance of attending to interpersonal relationships during tutoring. We then explain the core ideas of small group interactions, dialogue, relational interactions, care and belonging and provide a synthesis of these constructs. The literature synthesis presented is intended to be applied to research-based efforts aimed at supporting tutors working to increase their skills for cultivating strong interpersonal relationships and enacting equity oriented pedagogy. 

This paper shares a synthesis of the literature related to the application of a relationships-first approach to high-dosage math tutoring. In the context of our research, high-dosage tutoring is delivered multiple times per week during the school day by paraprofessionals who work with students in historically under-resourced schools. We apply a critical perspective to frame the importance of attending to interpersonal relationships during tutoring. We then explain the core ideas of small group interactions, dialogue, relational interactions, care and belonging and provide a synthesis of these constructs. The literature synthesis presented is intended to be applied to research-based efforts aimed at supporting tutors working to increase their skills for cultivating strong interpersonal relationships and enacting equity oriented pedagogy. 

Human tutoring interventions play a crucial role in supporting student learning, improving academic performance, and promoting personal growth. This paper focuses on analyzing mathematics tutoring discourse using talk moves—a framework of dialogue acts grounded in Accountable Talk theory. However, scaling the collection, annotation, and analysis of extensive tutoring dialogues to develop machine learning models is a challenging and resource-intensive task. To address this, we present SAGA22, a compact dataset, and explore various modeling strategies, including dialogue context, speaker information, pretraining datasets, and further fine-tuning. By leveraging existing datasets and models designed for classroom teaching, our results demonstrate that supplementary pretraining on classroom data enhances model performance in tutoring settings, particularly when incorporating longer context and speaker information. Additionally, we conduct extensive ablation studies to underscore the challenges in talk move modeling. 

Human tutoring interventions play a crucial role in supporting student learning, improving academic performance, and promoting personal growth. This paper focuses on analyzing mathematics tutoring discourse using talk moves—a framework of dialogue acts grounded in Accountable Talk theory. However, scaling the collection, annotation, and analysis of extensive tutoring dialogues to develop machine learning models is a challenging and resource-intensive task. To address this, we present SAGA22, a compact dataset, and explore various modeling strategies, including dialogue context, speaker information, pretraining datasets, and further fine-tuning. By leveraging existing datasets and models designed for classroom teaching, our results demonstrate that supplementary pretraining on classroom data enhances model performance in tutoring settings, particularly when incorporating longer context and speaker information. Additionally, we conduct extensive ablation studies to underscore the challenges in talk move modeling. 

Deep learning-based object detection algorithms enable the simultaneous classification and localization of any number of objects in image data. Many of these algorithms are capable of operating in real-time on high resolution images, attributing to their widespread usage across many fields. We present an end-to-end object detection pipeline designed for rare event searches for the Migdal effect, at real-time speeds, using high-resolution image data from the scientific CMOS camera readout of the MIGDAL experiment. The Migdal effect in nuclear scattering, critical for sub-GeV dark matter searches, has yet to be experimentally confirmed, making its detection a primary goal of the MIGDAL experiment. The Migdal effect forms a composite rare event signal topology consisting of an electronic and nuclear recoil sharing the same vertex. Crucially, both recoil species are commonly observed in isolation in the MIGDAL experiment, enabling us to train YOLOv8, a state-of-the-art object detection algorithm, on real data. Topologies indicative of the Migdal effect can then be identified in science data via pairs of neighboring or overlapping electron and nuclear recoils. Applying selections to real data that retain 99.7% signal acceptance in simulations, we demonstrate our pipeline to reduce a sample of 20 million recorded images to fewer than 1000 frames, thereby transforming a rare search into a much more manageable search. More broadly, we discuss the applicability of using object detection to enable data-driven machine learning training for other rare event search applications such as neutrinoless double beta decay searches and experiments imaging exotic nuclear decays. Published by the American Physical Society2025 

Abstract Waterfalls are among the fastest-eroding parts of river networks, but predicting natural waterfall retreat rates is difficult due to multiple processes that can drive waterfall erosion. We lack data on how waterfall height influences the mechanism and rate of upstream waterfall retreat. We addressed this knowledge gap with experiments testing the influence of drop height on waterfall retreat. Our experiments showed that shorter waterfalls retreat up to five times faster than taller waterfalls, when bedrock strength, sediment supply, and water discharge are constant. This retreat rate difference is due to a change in the erosion mechanism. Short waterfalls retreat by the formation of several small, rapidly eroding bedrock steps (i.e., cyclic steps), whereas tall waterfalls tend to form large bedrock plunge pools where lateral plunge pool erosion allows headwall undercutting and subsequent waterfall retreat. Because waterfall height can be partially set by the waterfall formation mechanism, our results highlight that the rate of waterfall retreat and subsequent landscape evolution can be modulated by the processes that form waterfalls. 

Transcripts of teaching episodes can be effective tools to understand discourse patterns in classroom instruction. According to most educational experts, sustained classroom discourse is a critical component of equitable, engaging, and rich learning environments for students. This paper describes the TalkMoves dataset, composed of 567 human annotated K-12 mathematics lesson transcripts (including entire lessons or portions of lessons) derived from video recordings. The set of transcripts primarily includes in-person lessons with whole-class discussions and/or small group work, as well as some online lessons. All of the transcripts are human-transcribed, segmented by the speaker (teacher or student), and annotated at the sentence level for ten discursive moves based on accountable talk theory. In addition, the transcripts include utterance-level information in the form of dialogue act labels based on the Switchboard Dialog Act Corpus. The dataset can be used by educators, policymakers, and researchers to understand the nature of teacher and student discourse in K-12 math classrooms. Portions of this dataset have been used to develop the TalkMoves application, which provides teachers with automated, immediate, and actionable feedback about their mathematics instruction.