arXiv subjects

Xiaoming Zhai

Publications and source records attributed to Xiaoming Zhai.

At least 37 records · Page 2Linked to original sources

arXiv2026

Science Literacy: Generative AI as Enabler of Coherence in the Teaching, Learning, and Assessment of Scientific Knowledge and Reasoning

This chapter examines the potential of generative AI in enhancing science literacy across the K-16+ grade span, including its benefits as well as the conceptual and practical challenges that doing so presents. It begins with a discussion of what defines science literacy in the era of AI, including how AI has changed science and the demand for future citizens to be scientifically literate when AI is applied in their careers and lives. The chapter further discusses why science literacy presents such a challenge in K-16+ educational settings. It then develops an argument for the type of architecture needed for AI to assist in solving the problem by bringing coherence to the teaching, learning, and assessment of science knowledge and reasoning. Components of this architecture are illustrated with respect to the AI tools and capabilities needed for design and implementation. The chapter concludes with a consideration of what has been learned regarding both science literacy and AI, as well as what remains to be learned, including the research and development (R&D) needed, and the generalizability of this science literacy case to other disciplinary learning and knowledge domains.

cs.CY↗

arXiv2026

Roots Beneath the Cut: Uncovering the Risk of Concept Revival in Pruning-Based Unlearning for Diffusion Models

Pruning-based unlearning has recently emerged as a fast, training-free, and data-independent approach to remove undesired concepts from diffusion models. It promises high efficiency and robustness, offering an attractive alternative to traditional fine-tuning or editing-based unlearning. However, in this paper we uncover a hidden danger behind this promising paradigm. We find that the locations of pruned weights, typically set to zero during unlearning, can act as side-channel signals that leak critical information about the erased concepts. To verify this vulnerability, we design a novel attack framework capable of reviving erased concepts from pruned diffusion models in a fully data-free and training-free manner. Our experiments confirm that pruning-based unlearning is not inherently secure, as erased concepts can be effectively revived without any additional data or retraining. Extensive experiments on diffusion-based unlearning based on concept related weights lead to the conclusion: once the critical concept-related weights in diffusion models are identified, our method can effectively recover the original concept regardless of how the weights are manipulated. Finally, we explore potential defense strategies and advocate safer pruning mechanisms that conceal pruning locations while preserving unlearning effectiveness, providing practical insights for designing more secure pruning-based unlearning frameworks.

cs.CV↗

arXiv2026

Transforming Science Learning Materials in the Era of Artificial Intelligence

The integration of artificial intelligence (AI) into science education is transforming the design and function of learning materials, offering new affordances for personalization, authenticity, and accessibility. This chapter examines how AI technologies are transforming science learning materials across six interrelated domains: 1) integrating AI into scientific practice, 2) enabling adaptive and personalized instruction, 3) facilitating interactive simulations, 4) generating multimodal content, 5) enhancing accessibility for diverse learners, and 6) promoting co-creation through AI-supported content development. These advancements enable learning materials to more accurately reflect contemporary scientific practice, catering to the diverse needs of students. For instance, AI support can enable students to engage in dynamic simulations, interact with real-time data, and explore science concepts through multimodal representations. Educators are increasingly collaborating with generative AI tools to develop timely and culturally responsive instructional resources. However, these innovations also raise critical ethical and pedagogical concerns, including issues of algorithmic bias, data privacy, transparency, and the need for human oversight. To ensure equitable and meaningful science learning, we emphasize the importance of designing AI-supported materials with careful attention to scientific integrity, inclusivity, and student agency. This chapter advocates for a responsible, ethical, and reflective approach to leveraging AI in science education, framing it as a catalyst for innovation while upholding core educational values.

cs.CY↗

arXiv2026

Charting the Future of AI-supported Science Education: A Human-Centered Vision

This concluding chapter explores how artificial intelligence (AI) is reshaping the purposes, practices, and outcomes of science education, and proposes a human-centered framework for its responsible integration. Drawing on insights from international collaborations and the Advancing AI in Science Education (AASE) committee, the chapter synthesizes developments across five dimensions: educational goals, instructional procedures, learning materials, assessment, and outcomes. We argue that AI offers transformative potential to enrich inquiry, personalize learning, and support teacher practice, but only when guided by Responsible and Ethical Principles (REP). The REP framework, emphasizing fairness, transparency, privacy, accountability, and respect for human values, anchors our vision for AI-supported science education. Key discussions include the redefinition of scientific literacy to encompass AI literacy, the evolving roles of teachers and learners in AI-supported classrooms, and the design of adaptive learning materials and assessments that preserve authenticity and integrity. We highlight both opportunities and risks, stressing the need for critical engagement with AI to avoid reinforcing inequities or undermining human agency. Ultimately, this chapter advances a vision in which science education prepares learners to act as ethical investigators and responsible citizens, ensuring that AI innovation aligns with human dignity, equity, and the broader goals of scientific literacy.

cs.CY↗

arXiv2026

Developing a Multi-Agent System to Generate Next Generation Science Assessments with Evidence-Centered Design

Contemporary science education reforms such as the Next Generation Science Standards (NGSS) demand assessments to understand students' ability to use science knowledge to solve problems and design solutions. To elicit such higher-order ability, educators need performance-based assessments, which are challenging to develop. One solution that has been broadly adopted is Evidence-Centered Design (ECD), which emphasizes interconnected models of the learner, evidence, and tasks. Although ECD provides a framework to safeguard assessment validity, its implementation requires diverse expertise (e.g., content and assessment), which is both costly and labor-intensive. To address this challenge, this study proposed integrating the ECD framework into Multi-Agent Systems (MAS) to generate NGSS-aligned assessment items automatically. This integrated MAS system ensembles multiple large language models with varying expertise, enabling the automation of complex, multi-stage item generation workflows traditionally performed by human experts. We examined the quality of AI-generated NGSS-aligned items and compared them with human-developed items across multiple dimensions of assessment design. Results showed that AI-generated items have overall comparable quality to human-developed items in terms of alignment with NGSS three-dimensional standards and cognitive demands. Divergent patterns also emerged: AI-generated items demonstrated a distinct strength in inclusivity, while also exhibiting limitations in clarity, conciseness, and multimodal design. AI- and human-developed items both showed weaknesses in evidence collectability and student interest alignment. These findings suggest that integrating ECD into MAS can support scalable and standards-aligned assessment design, while human expertise remains essential.

cs.CY↗

arXiv2026

Human-AI Collaborative Inductive Thematic Analysis: AI Guided Analysis and Human Interpretive Authority

The increasing use of generative artificial intelligence (GenAI) in qualitative research raises important questions about analytic practice and interpretive authority. This study examines how researchers interact with an Inductive Thematic Analysis GPT (ITA-GPT), a purpose-built AI tool designed to support inductive thematic analysis through structured, semi-automated prompts aligned with reflexive thematic analysis and verbatim coding principles. Guided by a Human-Artificial Intelligence Collaborative Inductive Thematic Analysis (HACITA) framework, the study focuses on analytic process rather than substantive findings. Three experienced qualitative researchers conducted ITA-GPT assisted analyses of interview transcripts from education research in the Ghanaian teacher education context. The tool supported familiarization, verbatim in vivo coding, gerund-based descriptive coding, and theme development, while enforcing trace to text integrity, coverage checks, and auditability. Data sources included interaction logs, AI-generated tables, researcher revisions, deletions, insertions, comments, and reflexive memos. Findings show that ITA-GPT functioned as a procedural scaffold that structured analytic workflow and enhanced transparency. However, interpretive authority remained with human researchers, who exercised judgment through recurrent analytic actions including modification, deletion, rejection, insertion, and commenting. The study demonstrates how inductive thematic analysis is enacted through responsible human AI collaboration.

cs.AI↗

arXiv2026

Q-realign: Piggybacking Realignment on Quantization for Safe and Efficient LLM Deployment

Public large language models (LLMs) are typically safety-aligned during pretraining, yet task-specific fine-tuning required for deployment often erodes this alignment and introduces safety risks. Existing defenses either embed safety recovery into fine-tuning or rely on fine-tuning-derived priors for post-hoc correction, leaving safety recovery tightly coupled with training and incurring high computational overhead and a complex workflow. To address these challenges, we propose \texttt{Q-realign}, a post-hoc defense method based on post-training quantization, guided by an analysis of representational structure. By reframing quantization as a dual-objective procedure for compression and safety, \texttt{Q-realign} decouples safety alignment from fine-tuning and naturally piggybacks into modern deployment pipelines. Experiments across multiple models and datasets demonstrate that our method substantially reduces unsafe behaviors while preserving task performance, with significant reductions in memory usage and GPU hours. Notably, our approach can recover the safety alignment of a fine-tuned 7B LLM on a single RTX 4090 within 40 minutes. Overall, our work provides a practical, turnkey solution for safety-aware deployment.

cs.LG↗

arXiv2026

Knowledge Distillation and Dataset Distillation of Large Language Models: Emerging Trends, Challenges, and Future Directions

The exponential growth of Large Language Models (LLMs) continues to highlight the need for efficient strategies to meet ever-expanding computational and data demands. This survey provides a comprehensive analysis of two complementary paradigms: Knowledge Distillation (KD) and Dataset Distillation (DD), both aimed at compressing LLMs while preserving their advanced reasoning capabilities and linguistic diversity. We first examine key methodologies in KD, such as task-specific alignment, rationale-based training, and multi-teacher frameworks, alongside DD techniques that synthesize compact, high-impact datasets through optimization-based gradient matching, latent space regularization, and generative synthesis. Building on these foundations, we explore how integrating KD and DD can produce more effective and scalable compression strategies. Together, these approaches address persistent challenges in model scalability, architectural heterogeneity, and the preservation of emergent LLM abilities. We further highlight applications across domains such as healthcare and education, where distillation enables efficient deployment without sacrificing performance. Despite substantial progress, open challenges remain in preserving emergent reasoning and linguistic diversity, enabling efficient adaptation to continually evolving teacher models and datasets, and establishing comprehensive evaluation protocols. By synthesizing methodological innovations, theoretical foundations, and practical insights, our survey charts a path toward sustainable, resource-efficient LLMs through the tighter integration of KD and DD principles.

cs.CL↗

arXiv2025

Report on the Scoping Workshop on AI in Science Education Research 2025

This report summarizes the outcomes of a two-day international scoping workshop on the role of artificial intelligence (AI) in science education research. As AI rapidly reshapes scientific practice, classroom learning, and research methods, the field faces both new opportunities and significant challenges. The report clarifies key AI concepts to reduce ambiguity and reviews evidence of how AI influences scientific work, teaching practices, and disciplinary learning. It identifies how AI intersects with major areas of science education research, including curriculum development, assessment, epistemic cognition, inclusion, and teacher professional development, highlighting cases where AI can support human reasoning and cases where it may introduce risks to equity or validity. The report also examines how AI is transforming methodological approaches across quantitative, qualitative, ethnographic, and design-based traditions, giving rise to hybrid forms of analysis that combine human and computational strengths. To guide responsible integration, a systems-thinking heuristic is introduced that helps researchers consider stakeholder needs, potential risks, and ethical constraints. The report concludes with actionable recommendations for training, infrastructure, and standards, along with guidance for funders, policymakers, professional organizations, and academic departments. The goal is to support principled and methodologically sound use of AI in science education research.

physics.ed-ph↗

arXiv2025

Human Experts' Evaluation of Generative AI for Contextualizing STEAM Education in the Global South

STEAM education in many parts of the Global South remains abstract and weakly connected to learners sociocultural realities. This study examines how human experts evaluate the capacity of Generative AI (GenAI) to contextualize STEAM instruction in these settings. Using a convergent mixed-methods design grounded in human-centered and culturally responsive pedagogy, four STEAM education experts reviewed standardized Ghana NaCCA lesson plans and GenAI-generated lessons created with a customized Culturally Responsive Lesson Planner (CRLP). Quantitative data were collected with a validated 25-item Culturally Responsive Pedagogy Rubric assessing bias awareness, cultural representation, contextual relevance, linguistic responsiveness, and teacher agency. Qualitative reflections provided additional insight into the pedagogical and cultural dynamics of each lesson. Findings show that GenAI, especially through the CRLP, improved connections between abstract standards and learners lived experiences. Teacher Agency was the strongest domain, while Cultural Representation was the weakest. CRLP-generated lessons were rated as more culturally grounded and pedagogically engaging. However, GenAI struggled to represent Ghana's cultural diversity, often producing surface-level references, especially in Mathematics and Computing. Experts stressed the need for teacher mediation, community input, and culturally informed refinement of AI outputs. Future work should involve classroom trials, broader expert participation, and fine-tuning with Indigenous corpora.

cs.CY↗

arXiv2025

Generative AI as a Learning Buddy and Teaching Assistant: Pre-service Teachers' Uses and Attitudes

This cross-sectional study investigates how preservice teachers in the Global South engage with Generative Artificial Intelligence across academic and instructional tasks while navigating infrastructural barriers such as limited internet access and high data costs. The study surveyed 167 preservice teachers from four teacher education institutions in Ghana. Descriptive statistics and inferential analyses, including multiple and ordinal logistic regressions, were used to examine patterns of GenAI use. Findings show that preservice teachers rely on GenAI as a learning companion for locating reading materials, accessing detailed content explanations, and identifying practical examples. They also use GenAI as a teaching assistant for tasks related to lesson preparation, including generating instructional resources, identifying assessment strategies, and developing lesson objectives. Usage patterns indicate that students in their third and fourth years have significantly higher frequencies of GenAI use compared to those in earlier years. Gender was not a significant predictor of GenAI adoption, in contrast to class level and age. Participants reported positive attitudes toward GenAI, noting that it supports autonomous learning and reduces dependence on peers and instructors for routine academic and teaching activities. However, challenges such as high data costs, occasional inaccuracies in GenAI outputs, and concerns about academic dishonesty were identified as factors that limit more frequent use. The study recommends the integration of GenAI literacy in teacher education programs, with a focus on ethical and responsible AI use to support equitable adoption in the Global South.

cs.HC↗

arXiv2025

Build AI Assistants using Large Language Models and Agents to Enhance the Engineering Education of Biomechanics

While large language models (LLMs) have demonstrated remarkable versatility across a wide range of general tasks, their effectiveness often diminishes in domain-specific applications due to inherent knowledge gaps. Moreover, their performance typically declines when addressing complex problems that require multi-step reasoning and analysis. In response to these challenges, we propose leveraging both LLMs and AI agents to develop education assistants aimed at enhancing undergraduate learning in biomechanics courses that focus on analyzing the force and moment in the musculoskeletal system of the human body. To achieve our goal, we construct a dual-module framework to enhance LLM performance in biomechanics educational tasks: 1) we apply Retrieval-Augmented Generation (RAG) to improve the specificity and logical consistency of LLM's responses to the conceptual true/false questions; 2) we build a Multi-Agent System (MAS) to solve calculation-oriented problems involving multi-step reasoning and code execution. Specifically, we evaluate the performance of several LLMs, i.e., Qwen-1.0-32B, Qwen-2.5-32B, and Llama-70B, on a biomechanics dataset comprising 100 true/false conceptual questions and problems requiring equation derivation and calculation. Our results demonstrate that RAG significantly enhances the performance and stability of LLMs in answering conceptual questions, surpassing those of vanilla models. On the other hand, the MAS constructed using multiple LLMs demonstrates its ability to perform multi-step reasoning, derive equations, execute code, and generate explainable solutions for tasks that require calculation. These findings demonstrate the potential of applying RAG and MAS to enhance LLM performance for specialized courses in engineering curricula, providing a promising direction for developing intelligent tutoring in engineering education.

cs.AI↗

arXiv2025

Generalizable and Efficient Automated Scoring with a Knowledge-Distilled Multi-Task Mixture-of-Experts

Automated scoring of written constructed responses typically relies on separate models per task, straining computational resources, storage, and maintenance in real-world education settings. We propose UniMoE-Guided, a knowledge-distilled multi-task Mixture-of-Experts (MoE) approach that transfers expertise from multiple task-specific large models (teachers) into a single compact, deployable model (student). The student combines (i) a shared encoder for cross-task representations, (ii) a gated MoE block that balances shared and task-specific processing, and (iii) lightweight task heads. Trained with both ground-truth labels and teacher guidance, the student matches strong task-specific models while being far more efficient to train, store, and deploy. Beyond efficiency, the MoE layer improves transfer and generalization: experts develop reusable skills that boost cross-task performance and enable rapid adaptation to new tasks with minimal additions and tuning. On nine NGSS-aligned science-reasoning tasks (seven for training/evaluation and two held out for adaptation), UniMoE-Guided attains performance comparable to per-task models while using $\sim$6$\times$ less storage than maintaining separate students, and $87\times$ less than the 20B-parameter teacher. The method offers a practical path toward scalable, reliable, and resource-efficient automated scoring for classroom and large-scale assessment systems.

cs.LG↗

arXiv2025

SketchMind: A Multi-Agent Cognitive Framework for Assessing Student-Drawn Scientific Sketches

Scientific sketches (e.g., models) offer a powerful lens into students' conceptual understanding, yet AI-powered automated assessment of such free-form, visually diverse artifacts remains a critical challenge. Existing solutions often treat sketch evaluation as either an image classification task or monolithic vision-language models, which lack interpretability, pedagogical alignment, and adaptability across cognitive levels. To address these limitations, we present SketchMind, a cognitively grounded, multi-agent framework for evaluating and improving student-drawn scientific sketches. SketchMind comprises modular agents responsible for rubric parsing, sketch perception, cognitive alignment, and iterative feedback with sketch modification, enabling personalized and transparent evaluation. We evaluate SketchMind on a curated dataset of 3,575 student-generated sketches across six science assessment items with different highest order of Bloom's level that require students to draw models to explain phenomena. Compared to baseline GPT-4o performance without SRG (average accuracy: 55.6%), and with SRG integration achieves 77.1% average accuracy (+21.4% average absolute gain). We also demonstrate that multi-agent orchestration with SRG enhances SketchMind performance, for example, GPT-4.1 gains an average 8.9% increase in sketch prediction accuracy, outperforming single-agent pipelines across all items. Human evaluators rated the feedback and co-created sketches generated by \textsc{SketchMind} with GPT-4.1, which achieved an average of 4.1 out of 5, significantly higher than those of baseline models (e.g., 2.3 for GPT-4o). Experts noted the system's potential to meaningfully support conceptual growth through guided revision. Our code and (pending approval) dataset will be released to support reproducibility and future research in AI-driven education.

cs.HC↗

arXiv2025

End-to-End On-Device Quantization-Aware Training for LLMs at Inference Cost

Quantization is an effective technique to reduce the deployment cost of large language models (LLMs), and post-training quantization (PTQ) has been widely studied due to its efficiency. However, existing PTQ methods are limited by their inability to fine-tune model parameters and often suffer significant accuracy loss in low-bit scenarios. Quantization-aware training (QAT) provides a more principled solution, but its reliance on backpropagation incurs prohibitive memory costs, limiting its practicality for LLM deployment. To address these challenges, we propose ZeroQAT, a zeroth-order optimization-based QAT framework that supports both weight and activation quantization. ZeroQAT leverages forward-only gradient estimation to eliminate backpropagation, substantially reducing computational and memory overhead while retaining the benefits of end-to-end optimization. We further introduce a lightweight variant of ZeroQAT for quantized fine-tuning, which freezes and pre-quantizes most parameters to further cut memory usage. Experiments show that ZeroQAT consistently outperforms representative PTQ and QAT baselines while requiring significantly less memory. For example, ZeroQAT enables fine-tuning of a 13B model at extremely low bit-widths (e.g., 2-4 bits) on a single 8GB GPU, and even allows fine-tuning a 6.7B model on a OnePlus 12 smartphone, demonstrating its practicality for end-to-end QAT on resource-limited edge devices.

cs.LG↗

arXiv2025

AutoSCORE: Enhancing Automated Scoring with Multi-Agent Large Language Models via Structured Component Recognition

Automated scoring plays a crucial role in education by reducing the reliance on human raters, offering scalable and immediate evaluation of student work. While large language models (LLMs) have shown strong potential in this task, their use as end-to-end raters faces challenges such as low accuracy, prompt sensitivity, limited interpretability, and rubric misalignment. These issues hinder the implementation of LLM-based automated scoring in assessment practice. To address the limitations, we propose AutoSCORE, a multi-agent LLM framework enhancing automated scoring via rubric-aligned Structured COmponent REcognition. With two agents, AutoSCORE first extracts rubric-relevant components from student responses and encodes them into a structured representation (i.e., Scoring Rubric Component Extraction Agent), which is then used to assign final scores (i.e., Scoring Agent). This design ensures that model reasoning follows a human-like grading process, enhancing interpretability and robustness. We evaluate AutoSCORE on four benchmark datasets from the ASAP benchmark, using both proprietary and open-source LLMs (GPT-4o, LLaMA-3.1-8B, and LLaMA-3.1-70B). Across diverse tasks and rubrics, AutoSCORE consistently improves scoring accuracy, human-machine agreement (QWK, correlations), and error metrics (MAE, RMSE) compared to single-agent baselines, with particularly strong benefits on complex, multi-dimensional rubrics, and especially large relative gains on smaller LLMs. These results demonstrate that structured component recognition combined with multi-agent design offers a scalable, reliable, and interpretable solution for automated scoring.

cs.CL↗

arXiv2025

Generative Large Language Models for Knowledge Representation: A Systematic Review of Concept Map Generation

The rise of generative large language models (LLMs) has opened new opportunities for automating knowledge representation through concept maps, a long-standing pedagogical tool valued for fostering meaningful learning and higher-order thinking. Traditional construction of concept maps is labor-intensive, requiring significant expertise and time, limiting their scalability in education. This review systematically synthesizes the emerging body of research on LLM-enabled concept map generation, focusing on two guiding questions: (a) What methods and technical features of LLMs are employed to construct concept maps? (b) What empirical evidence exists to validate their educational utility? Through a comprehensive search across major databases and AI-in-education conference proceedings, 28 studies meeting rigorous inclusion criteria were analyzed using thematic synthesis. Findings reveal six major methodological categories: human-in-the-loop systems, weakly supervised learning models, fine-tuned domain-specific LLMs, pre-trained LLMs with prompt engineering, hybrid systems integrating knowledge bases, and modular frameworks combining symbolic and statistical tools. Validation strategies ranged from quantitative metrics (precision, recall, F1-score, semantic similarity) to qualitative evaluations (expert review, learner feedback). Results indicate LLM-generated maps hold promise for scalable, adaptive, and pedagogically relevant knowledge visualization, though challenges remain regarding validity, interpretability, multilingual adaptability, and classroom integration. Future research should prioritize interdisciplinary co-design, empirical classroom trials, and alignment with instructional practices to realize their full educational potential.

cs.CY↗

arXiv2025

Rethinking the Potential of Layer Freezing for Efficient DNN Training

With the growing size of deep neural networks and datasets, the computational costs of training have significantly increased. The layer-freezing technique has recently attracted great attention as a promising method to effectively reduce the cost of network training. However, in traditional layer-freezing methods, frozen layers are still required for forward propagation to generate feature maps for unfrozen layers, limiting the reduction of computation costs. To overcome this, prior works proposed a hypothetical solution, which caches feature maps from frozen layers as a new dataset, allowing later layers to train directly on stored feature maps. While this approach appears to be straightforward, it presents several major challenges that are severely overlooked by prior literature, such as how to effectively apply augmentations to feature maps and the substantial storage overhead introduced. If these overlooked challenges are not addressed, the performance of the caching method will be severely impacted and even make it infeasible. This paper is the first to comprehensively explore these challenges and provides a systematic solution. To improve training accuracy, we propose \textit{similarity-aware channel augmentation}, which caches channels with high augmentation sensitivity with a minimum additional storage cost. To mitigate storage overhead, we incorporate lossy data compression into layer freezing and design a \textit{progressive compression} strategy, which increases compression rates as more layers are frozen, effectively reducing storage costs. Finally, our solution achieves significant reductions in training cost while maintaining model accuracy, with a minor time overhead. Additionally, we conduct a comprehensive evaluation of freezing and compression strategies, providing insights into optimizing their application for efficient DNN training.

cs.LG↗