arXiv subjects

Xinyun Wang

Publications and source records attributed to Xinyun Wang.

9 recordsLinked to original sources

arXiv2026

Ferroelectric brightening of spin forbidden dark excitons in a WSe2/hybrid perovskite heterostructure

Long-lived dark excitons in monolayer WSe2 present promising candidates for carrying spin and valley information, but their optical access and spin manipulation have conventionally required the use of strong external magnetic fields. Here, using a ferroelectric hybrid perovskite heterostructure, we leverage the ferroelectric proximity effect to break the WSe2's in-plane rotational symmetry and brighten the spin-forbidden dark excitons under zero magnetic field conditions. Furthermore, we show that the twist angle between the WSe2 and perovskite crystals controls the ferroelectric coupling strength and valley-contrasting polarization. Our proposed mechanism, supported by a four-band tight-binding model, suggests that the ferroelectric proximity effect induces an asymmetric intersublattice interaction, generating an effective in-plane spin-orbit coupling (SOC) field that rotates spin/valley polarization and brightens dark excitons. Our work establishes ferroelectric proximity coupling as an electrically reconfigurable, magnetic-field-free strategy for spin exciton control in two-dimensional semiconductors.

cond-mat.mtrl-sci↗

arXiv2026

EntSQL: A Benchmark for Grounding Text-to-SQL in Long-Context Enterprise Knowledge

Text-to-SQL enables natural language access to databases, and recent LLMs have substantially advanced its capabilities. Existing benchmarks such as Spider, BIRD, and Spider~2.0 evaluate schema generalization, large-scale databases, and realistic workflows, but largely overlook enterprise scenarios where SQL generation depends on private business knowledge, such as internal metrics, reporting conventions, and organizational rules. We introduce EntSQL, an enterprise-oriented Text-to-SQL benchmark for evaluating long-context grounding over proprietary business documents. EntSQL contains 1,066 aligned Chinese-English semantic examples across five business domains, with most examples requiring domain knowledge beyond the question and schema and involving complex SQL structures. On English inputs, the best evaluated system reaches only 15.9\% when long-form documents are provided, highlighting the difficulty of grounding SQL generation in enterprise knowledge.

cs.CL↗

arXiv2026

Reversible Diffusion Decoding for Diffusion Language Models

Diffusion language models enable parallel token generation through block-wise decoding, but their irreversible commitments can lead to stagnation, where the reverse diffusion process fails to make further progress under a suboptimal context.We propose Reversible Diffusion Decoding (RDD), a decoding framework that introduces reversibility into block-wise diffusion generation. RDD detects stagnation as a state-dependent failure of the reverse process and enables efficient backtracking to earlier blocks without recomputation via cached model states. To avoid repeated failure trajectories, RDD applies confidence-guided re-masking to selectively reinitialize uncertain tokens while preserving reliable context.This reversible formulation allows decoding to recover from early commitment errors while maintaining the parallel efficiency of diffusion-based generation. Experiments show that RDD improves generation robustness and quality over baselines with minimal computational overhead.

cs.CL↗

arXiv2025

AutoSynth: Automated Workflow Optimization for High-Quality Synthetic Dataset Generation via Monte Carlo Tree Search

Supervised fine-tuning (SFT) of large language models (LLMs) for specialized tasks requires high-quality datasets, but manual curation is prohibitively expensive. Synthetic data generation offers scalability, but its effectiveness relies on complex, multi-stage workflows, integrating prompt engineering and model orchestration. Existing automated workflow methods face a cold start problem: they require labeled datasets for reward modeling, which is especially problematic for subjective, open-ended tasks with no objective ground truth. We introduce AutoSynth, a framework that automates workflow discovery and optimization without reference datasets by reframing the problem as a Monte Carlo Tree Search guided by a novel dataset-free hybrid reward. This reward enables meta-learning through two LLM-as-judge components: one evaluates sample quality using dynamically generated task-specific metrics, and another assesses workflow code and prompt quality. Experiments on subjective educational tasks show that while expert-designed workflows achieve higher human preference rates (96-99% win rates vs. AutoSynth's 40-51%), models trained on AutoSynth-generated data dramatically outperform baselines (40-51% vs. 2-5%) and match or surpass expert workflows on certain metrics, suggesting discovery of quality dimensions beyond human intuition. These results are achieved while reducing human effort from 5-7 hours to just 30 minutes (>90% reduction). AutoSynth tackles the cold start issue in data-centric AI, offering a scalable, cost-effective method for subjective LLM tasks. Code: https://github.com/bisz9918-maker/AutoSynth.

cs.LG↗

arXiv2025

Cultivating Helpful, Personalized, and Creative AI Tutors: A Framework for Pedagogical Alignment using Reinforcement Learning

The integration of large language models (LLMs) into education presents unprecedented opportunities for scalable personalized learning. However, standard LLMs often function as generic information providers, lacking alignment with fundamental pedagogical principles such as helpfulness, student-centered personalization, and creativity cultivation. To bridge this gap, we propose EduAlign, a novel framework designed to guide LLMs toward becoming more effective and responsible educational assistants. EduAlign consists of two main stages. In the first stage, we curate a dataset of 8k educational interactions and annotate them-both manually and automatically-along three key educational dimensions: Helpfulness, Personalization, and Creativity (HPC). These annotations are used to train HPC-RM, a multi-dimensional reward model capable of accurately scoring LLM outputs according to these educational principles. We further evaluate the consistency and reliability of this reward model. In the second stage, we leverage HPC-RM as a reward signal to fine-tune a pre-trained LLM using Group Relative Policy Optimization (GRPO) on a set of 2k diverse prompts. We then assess the pre- and post-finetuning models on both educational and general-domain benchmarks across the three HPC dimensions. Experimental results demonstrate that the fine-tuned model exhibits significantly improved alignment with pedagogical helpfulness, personalization, and creativity stimulation. This study presents a scalable and effective approach to aligning LLMs with nuanced and desirable educational traits, paving the way for the development of more engaging, pedagogically aligned AI tutors.

cs.LG↗

arXiv2025

ELMES: An Automated Framework for Evaluating Large Language Models in Educational Scenarios

The emergence of Large Language Models (LLMs) presents transformative opportunities for education, generating numerous novel application scenarios. However, significant challenges remain: evaluation metrics vary substantially across different educational scenarios, while many emerging scenarios lack appropriate assessment metrics. Current benchmarks predominantly measure general intelligence rather than pedagogical capabilities. To address this gap, we introduce ELMES, an open-source automated evaluation framework specifically designed for assessing LLMs in educational settings. ELMES features a modular architecture that enables researchers to create dynamic, multi-agent dialogues through simple configuration files, facilitating flexible scenario design without requiring extensive programming expertise. The framework incorporates a hybrid evaluation engine that objectively quantifies traditionally subjective pedagogical metrics using an LLM-as-a-Judge methodology. We conduct systematic benchmarking of state-of-the-art LLMs across four critical educational scenarios: Knowledge Point Explanation, Guided Problem-Solving Teaching, Interdisciplinary Lesson Plan Generation, and Contextualized Question Generation, employing fine-grained metrics developed in collaboration with education specialists. Our results demonstrate distinct capability distributions among models, revealing context-specific strengths and limitations. ELMES provides educators and researchers with an accessible evaluation framework that significantly reduces adaptation barriers for diverse educational applications while advancing the practical implementation of LLMs in pedagogy. The framework is publicly available at \emph{https://github.com/sii-research/elmes.git}.

cs.CY↗

arXiv2025

Optical signatures of noncentrosymmetric structural distortion in altermagnetic MnTe

The hexagonal MnTe is a prime material candidate for altermagnets, an emerging class of magnetic compounds characterized by the nontrivial interplay of antiparallel spin arrangements with their underlying crystal structures. Recognizing precise knowledge of crystal symmetry as the cornerstone of the spin-group classification scheme, we report here a native inversion-symmetry-breaking structural distortion in this compound that has previously been overlooked. Through optical polarimetry experiments and first-principle calculations, we show that MnTe belongs to the noncentrosymmetric $D_{3h}$ group, effectively resolving key inconsistencies in the earlier interpretations of Raman spectroscopy data. Our finding impacts the symmetry analysis of MnTe within the altermagnetic class and sheds light on the mechanism of its magneto-controllable N\'eel order.

cond-mat.mtrl-sci↗

arXiv2022

Hausdorff Moment Transforms and Their Performance

Various methods have been proposed to approximate a solution to the truncated Hausdorff moment problem. In this paper, we establish a method of comparison for the performance of the approximations. Three ways of producing random moment sequences are discussed and applied. Also, some of the approximations have been rewritten as linear transforms, and detailed accuracy requirements are analyzed. Our finding shows that the performance of the approximations differs significantly in their convergence properties, accuracy, and numerical complexity and that the decay type of the moment sequence strongly affects the accuracy requirement.

math.NA↗

arXiv2018

Singlet superconductivity enhanced by charge order in nested twisted bilayer graphene Fermi surfaces

Using the continuum model for low energy non-interacting electronic structure of moiré van der Waals heterostructures developed by Bistritzer and MacDonald [1], we study the competition between spin, charge, and superconducting order in twisted bilayer graphene. Surprisingly, we find that for a range of small angles inclusive of the so-called magic angle, this model features robust Fermi pockets that preclude any Mott insulating phase at weak coupling. However, a Fermi surface reconstruction at $θ\gtrsim 1.2^{\circ}$ gives emergent van Hove singularities without any Fermi pockets. Using a hot-spot model for Fermi surface patches around these emergent saddle points, we develop a random-phase approximation from which we obtain a phase diagram very similar to that obtained recently by Isobe, Yuan, and Fu using the parquet renormalization group [2] but with additional insights. For example, our model shows strong nesting around time-reversal symmetric points at a moderate doping of $\sim 2 \times 10^{11} \text{cm}^{-2}$ away from the van Hove singularity. When this nesting dominates, we predict that charge-order enhances singlet superconductivity, while spin-order suppresses superconductivity. Our theory also provides additional possibilities for the case of unnested Fermi surfaces.

cond-mat.str-el↗