arXiv subjects

Rene F. Kizilcec

Publications and source records attributed to Rene F. Kizilcec.

At least 19 recordsLinked to original sources

arXiv2026

Think Before You Accept: Can Written Justification Reduce Uncritical Uptake of AI Writing Suggestions?

Generative AI can offer students useful feedback, but its value depends on judging which suggestions are accurate and relevant. Prior research shows that strategic friction during human-AI interactions can promote critical uptake, but how to effectively implement such friction in academic contexts remains unclear. We examine whether requiring students to justify decisions to accept or reject AI suggestions can mitigate uncritical uptake in academic writing. In a randomized experiment embedded in a course activity (N=129), students wrote a data analysis proposal, received mixed-quality AI revision suggestions, and decided whether to accept or reject them. Students required to provide written justifications were 24 percentage points less likely to adopt flawed suggestions (65% vs. 41%), with no reduction in acceptance of sound suggestions (81% vs. 86%). However, thematic analysis revealed superficial engagement in the justification task and gaps in metacognitive monitoring and domain knowledge.

cs.HC↗

arXiv2026

Data Annotation as Measurement

Modern AI systems depend on annotated data, but annotation is rarely treated as the act of measurement that it is. Instead, annotation quality is commonly reduced to agreement: if multiple annotators assign the same annotation to a data instance, the annotations are taken to be high-quality. Yet agreement does not establish whether annotations validly capture the underlying concept they are meant to represent. In this paper, we argue that data annotation should be understood as a measurement problem. Like other forms of measurement, annotation requires defining a concept, operationalizing it through an instrument, applying that instrument, and evaluating the reliability and validity of the resulting measurements. Drawing on a literature review of annotation quality research (N=132) and semi-structured interviews with annotation team members (N=10), we develop a framework for diagnosing and correcting annotation issues. First, we map key decision points across annotation processes - including task design, annotator management, quality assessment, quality improvement, and adjudication - that shape annotation outcomes. Second, we identify five distinct sources of annotation issues: error, ambiguity, impossibility, subjectivity, and annotator identity. Annotation problems that appear similar at the level of outcomes often require different process-level interventions based on their sources. Finally, we translate measurement theory into practical guidance for annotation teams, showing how assessments of reliability and validity can move beyond agreement alone. By reframing annotation as measurement, we offer a conceptual foundation for improving the quality of annotated data used in AI research and practice.

cs.CY↗

arXiv2026

Domain-Adapted Retrieval for In-Context Annotation of Pedagogical Dialogue Acts

Automated annotation of pedagogical dialogue is a high-stakes task where LLMs often fail without sufficient domain grounding. We present a domain-adapted RAG pipeline for tutoring move annotation. Rather than fine-tuning the generative model, we adapt retrieval by fine-tuning a lightweight embedding model on tutoring corpora and indexing dialogues at the utterance level to retrieve labeled few-shot demonstrations. Evaluated across two real tutoring dialogue datasets (TalkMoves and Eedi) and three LLM backbones (GPT-5.2, Claude Sonnet 4.6, Qwen3-32b), our best configuration achieves Cohen's $κ$ of 0.526-0.580 on TalkMoves and 0.659-0.743 on Eedi, substantially outperforming no-retrieval baselines ($κ= 0.275$-$0.413$ and $0.160$-$0.410$). An ablation study reveals that utterance-level indexing, rather than embedding quality alone, is the primary driver of these gains, with top-1 label match rates improving from 39.7\% to 62.0\% on TalkMoves and 52.9\% to 73.1\% on Eedi under domain-adapted retrieval. Retrieval also corrects systematic label biases present in zero-shot prompting and yields the largest improvements for rare and context-dependent labels. These findings suggest that adapting the retrieval component alone is a practical and effective path toward expert-level pedagogical dialogue annotation while keeping the generative model frozen.

cs.CL↗

arXiv2026

The Future of Feedback: How Can AI Help Transform Feedback to Be More Engaging, Effective, and Scalable?

With digital learning environments becoming more prevalent, the ease with which generative AI enables the scalable production of real-time, automated feedback holds the potential to reshape learning and teaching experiences. This meeting report synthesizes the interdisciplinary perspectives of 50 scholars from educational psychology, computer science, science education, and the learning sciences on the use of generative AI for feedback and its promises and risks in educational practice. We highlight points of convergence in the scholarship, identify areas of debate and unresolved challenges, and outline open questions and future directions for research and educational practice that emerged from structured small-group activities designed to bridge disciplinary barriers.

cs.CY↗

arXiv2026

Shiksha Copilot: Teacher-AI Collaboration for Curating and Customizing Lesson Plans in Low-Resource Schools

This study investigates Shiksha copilot, an AI-assisted lesson planning tool deployed in government schools across Karnataka, India. The system combined LLMs and human expertise through a structured process in which English and Kannada lesson plans were co-created by curators and AI; teachers then further customized these curated plans for their classrooms using their own expertise alongside AI support. Drawing on a large-scale mixed-methods study involving 1,043 teachers and 23 curators, we examine how educators collaborate with AI to generate context-sensitive lesson plans, assess the quality of AI-generated content, and analyze shifts in teaching practices within multilingual, low-resource environments. Our findings show that teachers used Shiksha copilot both to meet administrative documentation needs and to support their teaching. The tool eased bureaucratic workload, reduced lesson planning time, and lowered teaching-related stress, while promoting a shift toward activity-based pedagogy. However, systemic challenges such as staffing shortages and administrative demands constrained broader pedagogical change. We frame these findings through the lenses of teacher-AI collaboration and communities of practice to examine the effective integration of AI tools in teaching. Finally, we propose design directions for future teacher-centered EdTech, particularly in multilingual and Global South contexts.

cs.CY↗

arXiv2026

Optimizing LLM Annotation of Classroom Discourse through Multi-Agent Orchestration

Large language models (LLMs) are increasingly positioned as scalable tools for annotating educational data, including classroom discourse, interaction logs, and qualitative learning artifacts. Their ability to rapidly summarize instructional interactions and assign rubric-aligned labels has fueled optimism about reducing the cost and time associated with expert human annotation. However, growing evidence suggests that single-pass LLM outputs remain unreliable for high-stakes educational constructs that require contextual, pedagogical, or normative judgment, such as instructional intent or discourse moves. This tension between scale and validity sits at the core of contemporary education data science. In this work, we present and empirically evaluate a hierarchical, cost-aware orchestration framework for LLM-based annotation that improves reliability while explicitly modeling computational tradeoffs. Rather than treating annotation as a one-shot prediction problem, we conceptualize it as a multi-stage epistemic process comprising (1) an unverified single-pass annotation stage, in which models independently assign labels based on the rubric; (2) a self-verification stage, in which each model audits its own output against rubric definitions and revises its label if inconsistencies are detected; and (3) a disagreement-centric adjudication stage, in which an independent adjudicator model examines the verified labels and justifications and determines a final label in accordance with the rubric. This structure mirrors established human annotation workflows in educational research, where initial coding is followed by self-checking and expert resolution of disagreements.

cs.AI↗

arXiv2026

Protecting and Promoting Human Agency in Education in the Age of Artificial Intelligence

Human agency is crucial in education and increasingly challenged by the use of generative AI. This meeting report synthesizes interdisciplinary insights and conceptualizes four aspects that delineate human agency: human oversight, AI-human complementarity, AI competencies, and relational emergence. We explore practical dilemmas for protecting and promoting agency, focusing on normative constraints, transparency, and cognitive offloading, and highlight key tensions and implications to inform ethical and effective AI integration in education.

cs.HC↗

arXiv2026

The Digital Divide in Generative AI: Evidence from Large Language Model Use in College Admissions Essays

Large language models (LLMs) have become popular writing tools among students and may expand access to high-quality feedback for students with less access to traditional writing support. At the same time, LLMs may standardize student voice or invite overreliance. This study examines how adoption of LLM-assisted writing varies across socioeconomic groups and how it relates to outcomes in a high-stakes context: U.S. college admissions. We analyze a de-identified longitudinal dataset of applications to a selective university from 2020 to 2024 (N = 81,663). Estimating LLM use using a distribution-based detector trained on synthetic and historical essays, we tracked how student writing changed as LLM use proliferated, how adoption differed by socioeconomic status (SES), and whether potential benefits translated equitably into admissions outcomes. Using fee-waiver status as a proxy for SES, we observe post-2023 convergence in surface-level linguistic features, with the largest changes in fee-waived and rejected applicants. Estimated LLM use rose sharply in 2024 across all groups, with disproportionately larger increases among lower SES applicants, consistent with an access hypothesis in which LLMs substitute for scarce writing support. However, increased estimated LLM use was more strongly associated with declines in predicted admission probability for lower SES applicants than for higher SES applicants, even after controlling for academic credentials and stylometric features. These findings raise concerns about equity and the validity of essay-based evaluation in an era of AI-assisted writing and provide the first large-scale longitudinal evidence linking LLM adoption, linguistic change, and evaluative outcomes in college admissions.

cs.CY↗

arXiv2026

LLM Reasoning Predicts When Models Are Right: Evidence from Coding Classroom Discourse

Large Language Models (LLMs) are increasingly deployed to automatically label and analyze educational dialogue at scale, yet current pipelines lack reliable ways to detect when models are wrong. We investigate whether reasoning generated by LLMs can be used to predict the correctness of a model's own predictions. We analyze 30,300 teacher utterances from classroom dialogue, each labeled by multiple state-of-the-art LLMs with an instructional move construct and an accompanying reasoning. Using human-verified ground-truth labels, we frame the task as predicting whether a model's assigned label for a given utterance is correct. We encode LLM reasoning using Term Frequency-Inverse Document Frequency (TF-IDF) and evaluate five supervised classifiers. A Random Forest classifier achieves an F1 score of 0.83 (Recall = 0.854), successfully identifying most incorrect predictions and outperforming baselines. Training specialist detectors for specific instructional move constructs further improves performance on difficult constructs, indicating that error detection benefits from construct-specific linguistic cues. Using the Linguistic Inquiry and Word Count (LIWC) framework, we examine four linguistic markers of correctness: Causation, Differentiation, Tentativeness, and Insight. Correct predictions exhibit grounded causal language (e.g., because, therefore), while incorrect reasoning is substantially more likely to rely on epistemic hedging (e.g., might, could) and performative metacognition (e.g., think, realize). Syntactic complexity does not distinguish correct from incorrect reasoning, and longer reasoning is not more reliable. These findings demonstrate that reasoning-based error detection offers a practical and scalable approach to quality control in automated educational dialogue analysis.

cs.CL↗

arXiv2026

AI Annotation Orchestration: Evaluating LLM verifiers to Improve the Quality of LLM Annotations in Learning Analytics

Large Language Models (LLMs) are increasingly used to annotate learning interactions, yet concerns about reliability limit their utility. We test whether verification-oriented orchestration-prompting models to check their own labels (self-verification) or audit one another (cross-verification)-improves qualitative coding of tutoring discourse. Using transcripts from 30 one-to-one math sessions, we compare three production LLMs (GPT, Claude, Gemini) under three conditions: unverified annotation, self-verification, and cross-verification across all orchestration configurations. Outputs are benchmarked against a blinded, disagreement-focused human adjudication using Cohen's kappa. Overall, orchestration yields a 58 percent improvement in kappa. Self-verification nearly doubles agreement relative to unverified baselines, with the largest gains for challenging tutor moves. Cross-verification achieves a 37 percent improvement on average, with pair- and construct-dependent effects: some verifier-annotator pairs exceed self-verification, while others reduce alignment, reflecting differences in verifier strictness. We contribute: (1) a flexible orchestration framework instantiating control, self-, and cross-verification; (2) an empirical comparison across frontier LLMs on authentic tutoring data with blinded human "gold" labels; and (3) a concise notation, verifier(annotator) (e.g., Gemini(GPT) or Claude(Claude)), to standardize reporting and make directional effects explicit for replication. Results position verification as a principled design lever for reliable, scalable LLM-assisted annotation in Learning Analytics.

cs.AI↗

arXiv2026

Codebook-Injected Dialogue Segmentation for Multi-Utterance Constructs Annotation: LLM-Assisted and Gold-Label-Free Evaluation

Dialogue Act (DA) annotation typically treats communicative or pedagogical intent as localized to individual utterances or turns. This leads annotators to agree on the underlying action while disagreeing on segment boundaries, reducing apparent reliability. We propose codebook-injected segmentation, which conditions boundary decisions on downstream annotation criteria, and evaluate LLM-based segmenters against standard and retrieval-augmented baselines. To assess these without gold labels, we introduce evaluation metrics for span consistency, distinctiveness, and human-AI distributional agreement. We found DA-awareness produces segments that are internally more consistent than text-only baselines. While LLMs excel at creating construct-consistent spans, coherence-based baselines remain superior at detecting global shifts in dialogue flow. Across two datasets, no single segmenter dominates. Improvements in within-segment coherence frequently trade off against boundary distinctiveness and human-AI distributional agreement. These results highlight segmentation as a consequential design choice that should be optimized for downstream objectives rather than a single performance score.

cs.CL↗

arXiv2025

How well do Large Language Models Recognize Instructional Moves? Establishing Baselines for Foundation Models in Educational Discourse

Large language models (LLMs) are increasingly adopted in educational technologies for a variety of tasks, from generating instructional materials and assisting with assessment design to tutoring. While prior work has investigated how models can be adapted or optimized for specific tasks, far less is known about how well LLMs perform at interpreting authentic educational scenarios without significant customization. As LLM-based systems become widely adopted by learners and educators in everyday academic contexts, understanding their out-of-the-box capabilities is increasingly important for setting expectations and benchmarking. We compared six LLMs to estimate their baseline performance on a simple but important task: classifying instructional moves in authentic classroom transcripts. We evaluated typical prompting methods: zero-shot, one-shot, and few-shot prompting. We found that while zero-shot performance was moderate, providing comprehensive examples (few-shot prompting) significantly improved performance for state-of-the-art models, with the strongest configuration reaching Cohen's Kappa = 0.58 against expert-coded annotations. At the same time, improvements were neither uniform nor complete: performance varied considerably by instructional move, and higher recall frequently came at the cost of increased false positives. Overall, these findings indicate that foundation models demonstrate meaningful yet limited capacity to interpret instructional discourse, with prompt design helping to surface capability but not eliminating fundamental reliability constraints.

cs.CL↗

arXiv2025

Investigating Student Interaction Patterns with Large Language Model-Powered Course Assistants in Computer Science Courses

Providing students with flexible and timely academic support is a challenge at most colleges and universities, leaving many students without help outside scheduled hours. Large language models (LLMs) are promising for bridging this gap, but interactions between students and LLMs are rarely overseen by educators. We developed and studied an LLM-powered course assistant deployed across multiple computer science courses to characterize real-world use and understand pedagogical implications. By Spring 2024, our system had been deployed to approximately 2,000 students across six courses at three institutions. Analysis of the interaction data shows that usage remains strong in the evenings and nights and is higher in introductory courses, indicating that our system helps address temporal support gaps and novice learner needs. We sampled 200 conversations per course for manual annotation: most sampled responses were judged correct and helpful, with a small share unhelpful or erroneous; few responses included dedicated examples. We also examined an inquiry-based learning strategy: only around 11% of sampled conversations contained LLM-generated follow-up questions, which were often ignored by students in advanced courses. A Bloom's taxonomy analysis reveals that current LLM capabilities are limited in generating higher-order cognitive questions. These patterns suggest opportunities for pedagogically oriented LLM-based educational systems and greater educator involvement in configuring prompts, content, and policies.

cs.CY↗

arXiv2025

When the Past Misleads: Rethinking Training Data Expansion Under Temporal Distribution Shifts

Predictive models are typically trained on historical data to predict future outcomes. While it is commonly assumed that training on more historical data would improve model performance and robustness, data distribution shifts over time may undermine these benefits. This study examines how expanding historical data training windows under covariate shifts (changes in feature distributions) and concept shifts (changes in feature-outcome relationships) affects the performance and algorithmic fairness of predictive models. First, we perform a simulation study to explore scenarios with varying degrees of covariate and concept shifts in training data. Absent distribution shifts, we observe performance gains from longer training windows though they reach a plateau quickly; in the presence of concept shift, performance may actually decline. Covariate shifts alone do not significantly affect model performance, but may complicate the impact of concept shifts. In terms of fairness, models produce more biased predictions when the magnitude of concept shifts differs across sociodemographic groups; for intersectional groups, these effects are more complex and not simply additive. Second, we conduct an empirical case study of student retention prediction, a common machine learning application in education, using 12 years of student records from 23 minority-serving community colleges in the United States. We find concept shifts to be a key contributor to performance degradation when expanding the training window. Moreover, model fairness is compromised when marginalized populations have distinct data distribution shift patterns from their peers. Overall, our findings caution against conventional wisdom that "more data is better" and underscore the importance of using historical data judiciously, especially when it may be subject to data distribution shifts, to improve model performance and fairness.

cs.CY↗

arXiv2025

A Framework for Auditing Chatbots for Dialect-Based Quality-of-Service Harms

Increasingly, individuals who engage in online activities are expected to interact with large language model (LLM)-based chatbots. Prior work has shown that LLMs can display dialect bias, which occurs when they produce harmful responses when prompted with text written in minoritized dialects. However, whether and how this bias propagates to systems built on top of LLMs, such as chatbots, is still unclear. We conduct a review of existing approaches for auditing LLMs for dialect bias and show that they cannot be straightforwardly adapted to audit LLM-based chatbots due to issues of substantive and ecological validity. To address this, we present a framework for auditing LLM-based chatbots for dialect bias by measuring the extent to which they produce quality-of-service harms, which occur when systems do not work equally well for different people. Our framework has three key characteristics that make it useful in practice. First, by leveraging dynamically generated instead of pre-existing text, our framework enables testing over any dialect, facilitates multi-turn conversations, and represents how users are likely to interact with chatbots in the real world. Second, by measuring quality-of-service harms, our framework aligns audit results with the real-world outcomes of chatbot use. Third, our framework requires only query access to an LLM-based chatbot, meaning that it can be leveraged equally effectively by internal auditors, external auditors, and even individual users in order to promote accountability. To demonstrate the efficacy of our framework, we conduct a case study audit of Amazon Rufus, a widely-used LLM-based chatbot in the customer service domain. Our results reveal that Rufus produces lower-quality responses to prompts written in minoritized English dialects, and that these quality-of-service harms are exacerbated by the presence of typos in prompts.

cs.CY↗

arXiv2025

Benchmarking Generative AI for Scoring Medical Student Interviews in Objective Structured Clinical Examinations (OSCEs)

Objective Structured Clinical Examinations (OSCEs) are widely used to assess medical students' communication skills, but scoring interview-based assessments is time-consuming and potentially subject to human bias. This study explored the potential of large language models (LLMs) to automate OSCE evaluations using the Master Interview Rating Scale (MIRS). We compared the performance of four state-of-the-art LLMs (GPT-4o, Claude 3.5, Llama 3.1, and Gemini 1.5 Pro) in evaluating OSCE transcripts across all 28 items of the MIRS under the conditions of zero-shot, chain-of-thought (CoT), few-shot, and multi-step prompting. The models were benchmarked against a dataset of 10 OSCE cases with 174 expert consensus scores available. Model performance was measured using three accuracy metrics (exact, off-by-one, thresholded). Averaging across all MIRS items and OSCE cases, LLMs performed with low exact accuracy (0.27 to 0.44), and moderate to high off-by-one accuracy (0.67 to 0.87) and thresholded accuracy (0.75 to 0.88). A zero temperature parameter ensured high intra-rater reliability (α = 0.98 for GPT-4o). CoT, few-shot, and multi-step techniques proved valuable when tailored to specific assessment items. The performance was consistent across MIRS items, independent of encounter phases and communication domains. We demonstrated the feasibility of AI-assisted OSCE evaluation and provided benchmarking of multiple LLMs across multiple prompt techniques. Our work provides a baseline performance assessment for LLMs that lays a foundation for future research into automated assessment of clinical communication skills.

cs.CL↗

arXiv2025

"Don't Forget the Teachers": Towards an Educator-Centered Understanding of Harms from Large Language Models in Education

Education technologies (edtech) are increasingly incorporating new features built on large language models (LLMs), with the goals of enriching the processes of teaching and learning and ultimately improving learning outcomes. However, the potential downstream impacts of LLM-based edtech remain understudied. Prior attempts to map the risks of LLMs have not been tailored to education specifically, even though it is a unique domain in many respects: from its population (students are often children, who can be especially impacted by technology) to its goals (providing the correct answer may be less important for learners than understanding how to arrive at an answer) to its implications for higher-order skills that generalize across contexts (e.g., critical thinking and collaboration). We conducted semi-structured interviews with six edtech providers representing leaders in the K-12 space, as well as a diverse group of 23 educators with varying levels of experience with LLM-based edtech. Through a thematic analysis, we explored how each group is anticipating, observing, and accounting for potential harms from LLMs in education. We find that, while edtech providers focus primarily on mitigating technical harms, i.e., those that can be measured based solely on LLM outputs themselves, educators are more concerned about harms that result from the broader impacts of LLMs, i.e., those that require observation of interactions between students, educators, school systems, and edtech to measure. Overall, we (1) develop an education-specific overview of potential harms from LLMs, (2) highlight gaps between conceptions of harm by edtech providers and those by educators, and (3) make recommendations to facilitate the centering of educators in the design and development of edtech tools.

cs.CY↗

arXiv2024

Reflection on Purpose Changes Students' Academic Interests: A Scalable Intervention in an Online Course Catalog

College students routinely use online course catalogs to explore a variety of academic offerings. Course catalogs may therefore be an effective place to encourage reflection on academic choices and interests. To test this, we embedded a psychological intervention in an online course catalog to encourage students to reflect on their purpose during course exploration. Results of a randomized field experiment with over 4,000 students at a large U.S. university show that a purpose intervention increased students' cognitive engagement in describing their interests, but reduced search activities. Students became more interested in courses related to creative arts and social change, but less in computer and data science. The findings demonstrate the malleability of students' interests during course exploration and suggest practical strategies to support purpose reflection and guide students toward deliberate exploration of their interests in higher education.

cs.CY↗