Search arXiv⌕ Search

arXiv subjects

Min Zhang

Publications and source records attributed to Min Zhang.

At least 73 records · Page 4Linked to original sources
arXiv2026

Benchmarking LLM-as-a-Judge for Long-Form Output Evaluation

As large language models (LLMs) are increasingly used for long-form generation, reliably evaluating long-form outputs has become a critical challenge. LLM-as-a-judge offers a scalable alternative to human evaluation, yet its reliability in long-form output evaluation remains underexamined: existing meta-evaluation benchmarks focus mainly on short-form outputs. Compared with short-form evaluation, long-form evaluation is not merely a matter of output length; it often requires judges to make more complex document-level assessments of overall organization, task-relevant coverage and depth, cross-section consistency, and scenario-specific quality criteria. In this work, we introduce LongJudgeBench, a comprehensive benchmark for evaluating LLM judges on long-form outputs across diverse real-world scenarios and judging protocols. We systematically evaluate a broad range of LLM judges, covering multiple base models and judging settings. Our results reveal a substantial reliability gap: current LLM judges remain unstable across scenarios, and rubrics or references are helpful but not always sufficient. We hope LongJudgeBench will support future research on more robust, context-aware, and human-aligned LLM-as-a-judge methods. Our code is available at https://github.com/cjj826/LongJudgeBench.

cs.CL↗
arXiv2026

Integrating LLM and Diffusion-Based Agents for Social Simulation

Large language models (LLMs) offer strong semantic reasoning capabilities for user modeling, but applying LLM-based simulation to an entire social network is computationally expensive and often unreliable for users with sparse behavioral histories. Meanwhile, conventional information diffusion models efficiently exploit historical propagation patterns and social structures, but provide limited understanding of item content and user-item semantic compatibility. We propose HySID, a hybrid framework for individual-level information adoption prediction that combines semantic reasoning with structural diffusion. HySID first analyzes the historical user-relation graph to adaptively select a small set of structurally informative core users. It then applies LLM-based simulation to estimate the engagement of these users and converts the judgments into a diffusion-compatible seed. Finally, a plug-in diffusion backbone propagates this seed through historical interaction structures to rank potential adopters across the full user population. This design enables LLMs to focus on users for whom semantic reasoning is most informative while allowing neural diffusion models to generalize the evidence to users that are not explicitly simulated. Experiments on three real-world datasets from Weibo, Zhihu, and KuaiRand show that HySID consistently improves four diffusion backbones in Recall and NDCG, while outperforming full-population LLM simulation baselines. At the same time, selective simulation reduces LLM inference cost by approximately 83\% to 99.9\%, demonstrating that HySID provides an effective and computationally practical approach to scalable information adoption prediction.

cs.CY↗
arXiv2026

SMRC: Aligning Large Language Models with Student Reasoning for Mathematical Error Correction

Large language models (LLMs) often make reasoning errors when solving mathematical problems, and how to automatically detect and correct these errors has become an important research direction. However, existing approaches \textit{mainly focus on self-correction within the model}, which falls short of the "teacher-style" correction required in educational settings, \textit{i.e.}, systematically guiding and revising a student' s problem-solving process. To address this gap, we propose \texttt{SMRC} (\textit{\underline{S}tudent \underline{M}athematical \underline{R}easoning \underline{C}orrection}), a novel method that aligns LLMs with student reasoning. Specifically, \texttt{SMRC} formulates student reasoning as a multi-step sequential decision problem and introduces Monte Carlo Tree Search (MCTS) to explore optimal correction paths. To reduce the cost of the annotating process-level rewards, we leverage breadth-first search (BFS) guided by LLMs and final-answer evaluation to generate reward signals, which are then distributed across intermediate reasoning steps via a back-propagation mechanism, enabling fine-grained process supervision. Additionally, we construct a benchmark for high school mathematics, MSEB (Multi-Solution Error Benchmark), consisting of 158 instances that include problem statements, student solutions, and correct reasoning steps. We further propose a dual evaluation protocol centered on \textbf{solution accuracy} and \textbf{correct-step retention}, offering a comprehensive measure of educational applicability. Experiments demonstrate that \texttt{SMRC} significantly outperforms existing methods on two public datasets (ProcessBench and MR-GSM8K) and our MSEB in terms of effectiveness and overall performance. The code are available at https://github.com/ECNU-RAIL/SMRC-EMNLP2026.

cs.CL↗
arXiv2026

On Piatetski-Shapiro primes from almost primes

Denote by $\mathcal{P}_r$ an almost-prime with at most $r$ prime factors, counted according to multiplicity. In this manuscript, it is established that, for any fixed $0.98353<γ<1$, there exist infinitely many primes of the form $p=[n^{1/γ}]$, where $n$ is an almost-prime $\mathcal{P}_7$. This result constitutes an improvement upon the previous result of Baker, Banks, Guo and Yeager [1], who showed that there exist infinitely many primes $p$ such that $p=[n^{1/γ}]$ with $n\in\mathcal{P}_8$ for $γ$ near to one.

math.NT↗
arXiv2026

EduDial: Constructing a Large-scale Multi-turn Teacher-Student Dialogue Corpus

Recently, several multi-turn dialogue benchmarks have been proposed to evaluate the conversational abilities of large language models (LLMs). As LLMs are increasingly recognized as a key technology for advancing intelligent education, owing to their ability to deeply understand instructional contexts and provide personalized guidance, the construction of dedicated teacher-student dialogue benchmarks has become particularly important. To this end, we present EduDial, a comprehensive multi-turn teacher-student dialogue dataset. EduDial covers 345 core knowledge points and consists of 34,250 dialogue sessions generated through interactions between teacher and student agents. Its design is guided by Bloom's taxonomy of educational objectives and incorporates ten questioning strategies, including situational questioning, zone of proximal development (ZPD) questioning, and metacognitive questioning-thus better capturing authentic classroom interactions. Furthermore, we design differentiated teaching strategies for students at different cognitive levels, thereby providing more targeted teaching guidance. Building on EduDial, we further develop EduDial-LLM 32B via training and propose an 11-dimensional evaluation framework that systematically measures the teaching abilities of LLMs, encompassing both overall teaching quality and content quality. Experiments on 17 mainstream LLMs reveal that most models struggle in student-centered teaching scenarios, whereas our EduDial-LLM achieves significant gains, consistently outperforming all baselines across all metrics. The code is available at https://github.com/Mind-Lab-ECNU/EduDial/tree/main.

cs.CL↗
arXiv2026

UCO: A Multi-Turn Interactive Reinforcement Learning Method for Adaptive Teaching with Large Language Models

Large language models (LLMs) are shifting from answer providers to intelligent tutors in educational settings, yet current supervised fine-tuning methods only learn surface teaching patterns without dynamic adaptation capabilities. Recent reinforcement learning approaches address this limitation but face two critical challenges. First, they evaluate teaching effectiveness solely based on whether students produce correct outputs, unable to distinguish whether students genuinely understand or echo teacher-provided answers during interaction. Second, they cannot perceive students' evolving cognitive states in real time through interactive dialogue, thus failing to adapt teaching strategies to match students' cognitive levels dynamically. We propose the Unidirectional Cognitive Optimization (UCO) method to address these challenges. UCO uses a multi-turn interactive reinforcement learning paradigm where the innovation lies in two synergistic reward functions: the Progress Reward captures students' cognitive advancement, evaluating whether students truly transition from confusion to comprehension, while the Scaffold Reward dynamically identifies each student's Zone of Proximal Development (ZPD), encouraging teachers to maintain productive teaching within this zone. We evaluate UCO by comparing it against 11 baseline models on BigMath and MathTutorBench benchmarks. Experimental results demonstrate that our UCO model outperforms all models of equivalent scale and achieves performance comparable to advanced closed-source models. The code and data are available at https://github.com/Mind-Lab-ECNU/UCO.

cs.AI↗
arXiv2026

EduAgentQG: Multi-Agent Personalized Mathematics Question Generation with Explicit Diversity and Objective-Aware Evaluation

In intelligent education, personalized mathematics question generation aims to produce mathematics questions that satisfy educational requirements while supporting adaptive assessment and learning. Existing LLM-based single-agent and multi-agent methods improve generation flexibility, but they still tend to rely on aggregated feedback or model randomness, making it difficult to jointly ensure dimension-wise objective alignment and controllable diversity. To address these challenges, we propose EduAgentQG, a multi-agent collaborative framework for personalized mathematics question generation with explicit diversity and objective-aware evaluation. EduAgentQG organizes question generation as a closed-loop process of planning, writing, evaluation, refinement, and checking: structured generation plans and multiple generation directions guide candidate generation, while fine-grained evaluation verifies logical correctness, solvability, and objective alignment in knowledge concepts, difficulty, grade level, and core competencies. We first construct a mathematics question generation benchmark containing 10,273 questions across Grades 1-9, covering 634 knowledge concepts, 16 core competencies, and three difficulty levels; for evaluation, it is organized into two subsets: MathChoice, with 489 educational objectives for multiple-choice question generation, and MathBlank, with 500 educational objectives for fill-in-the-blank question generation. Experiments show that EduAgentQG consistently outperforms COT, COT$_N$, ReAct, and EQPR in diversity, Objective Consistency, and Win Rate.

cs.CY↗
arXiv2026

OmniPhys: A Unified Multimodal Benchmark for Physics Understanding and Generation from Chinese Educational Corpora

Multimodal Large Language Models (MLLMs) have demonstrated strong abilities in solving diverse visual and textual reasoning tasks. However, their development in the physics domain is significantly hindered by the lack of a comprehensive benchmark. To fill this gap, we introduce OmniPhys, a large-scale benchmark for multimodal physics understanding and reasoning, covering middle school through university-level problems from Chinese Educational Corpora. OmniPhys consists of 15,246 questions and 19,850 images, accompanied by detailed annotations that support fine-grained analysis of reasoning processes and knowledge usage. Beyond conventional evaluation, OmniPhys is a benchmark that systematically evaluates multimodal outputs in the physics domain, including models' ability to generate structured physics diagrams, which constitute a fundamental component of authentic physics problem solving. Extensive evaluations reveal critical gaps in the capabilities of current MLLMs, especially in complex reasoning and visual generation. To address this, we release OmniPhys to serve as a foundational resource for advancing multimodal intelligence in physics and scientific domains. Codes and data are available at https://github.com/ECNU-RAIL/OmniPhys-EMNLP2026.

cs.CL↗
arXiv2026

CASTLE: A Comprehensive Benchmark for Evaluating Student-Tailored Personalized Safety in Large Language Models

Large language models (LLMs) have advanced the development of personalized learning in education. However, their inherent generation mechanisms often produce homogeneous responses to identical prompts. This one-size-fits-all mechanism overlooks the substantial heterogeneity in students cognitive and psychological, thereby posing potential safety risks to vulnerable groups. Existing safety evaluations primarily rely on context-independent metrics such as factual accuracy, bias, or toxicity, which fail to capture the divergent harms that the same response might cause across different student attributes. To address this gap, we propose the concept of Student-Tailored Personalized Safety and construct CASTLE based on educational theories. This benchmark covers 15 educational safety risks and 14 student attributes, comprising 92,908 bilingual scenarios. We further design three evaluation metrics: Risk Sensitivity, measuring the model ability to detect risks; Emotional Empathy, evaluating the model capacity to recognize student states; and Student Alignment, assessing the match between model responses and student attributes. Experiments on 18 SOTA LLMs demonstrate that CASTLE poses a significant challenge: all models scored below an average safety rating of 2.3 out of 5, indicating substantial deficiencies in personalized safety assurance.

cs.CL↗
arXiv2026

qshap: Fast Shapley Decomposition of $R^2$ for Gradient-Boosted Trees

Numerous methods have been developed to quantify feature attributions in individual predictions for tree ensembles. However, many applications require global measures of feature contributions to overall model performance. Although local attribution scores can be aggregated to characterize feature importance, such summaries do not directly decompose measures of predictive performance, such as $R^2$. This article introduces qshap, available in both R and Python, which provides Shapley decomposition of $R^2$ values for gradient-boosted decision trees (GBDTs) to quantify feature-specific contributions to model performance. By decomposing the quadratic loss of individual observations, qshap provides flexible tools to explore the importance of individual features and observations. qshap currently supports widely used GBDT implementations, including xgboost, lightgbm, and catboost, through a unified tree representation and efficient C++ backends. Its modular design can accommodate other GBDT implementations built from binary decision trees. In addition, we introduce a specialized backend for oblivious trees that exploits their symmetric structure to substantially accelerate computation.

stat.ML↗
arXiv2026

Symmetric Taylor--Hood elements for the linear stress gradient problem

We develop mixed finite element methods for the linear stress-gradient elasticity problem based on symmetric Taylor--Hood elements. We establish the stability of the symmetric Taylor--Hood pair on simplicial meshes in both two and three dimensions, thereby resolving the stability left open by Brezzi, Fortin, and Marini in 1993. Combing the Nitsche's method, we further construct finite element schemes for linear stress problem in general boundary condition. Numerical experiments confirm the theoretical results.

math.NA↗
arXiv2026

SLAI T-Rex: Full-Parameter Post-training of the DeepSeek-V4 Family on Ascend SuperPOD

Full-parameter post-training of trillion-parameter-scale MoE models introduces substantial system-level challenges for large-scale distributed training, including severe memory pressure, non-overlapped communication overhead, and inefficient kernel execution. While most large-scale LLM training systems are built around GPU-based clusters, this report presents an end-to-end optimization practice on the Ascend NPU SuperPOD. Using the DeepSeek-V4 model family as the target workload, we develop a hierarchical optimization framework spanning model-level parallelism, computation-communication orchestration, and low-level kernel execution. The resulting system achieves 34.22% Model FLOPs Utilization (MFU) with a 2.93x improvement over the open-source baseline recipe while maintaining training stability. Building on this optimized infrastructure, we further establish a CPT and SFT workflow for complex Operations Research (OR) tasks. We refer to the integrated framework as SLAI T-Rex. Using DeepSeek-V4-Flash, we develop OR-oriented CPT and SFT data pipelines that combine collected domain resources with solver-verified synthetic optimization documents. The resulting dataset contains 10K high-quality SFT samples spanning four task categories and three problem representations. The specialized model achieves the highest average zero-shot Pass@1 score among the evaluated models, reaching 71.81% and outperforming GPT-5.4-Mini and the base DeepSeek-V4-Flash model by 3.98 and 11.27 percentage points, respectively. Overall, this work demonstrates a full-stack pathway from efficient trillion-parameter model post-training on Ascend infra to domain-specialized Flash models for solver-grounded mathematical modeling, advancing frontier-model systems for complex reasoning.

cs.CL↗
arXiv2026

OmniHandwritingOCR: A Diagnostic Benchmark for Evaluating Multimodal LLMs in Handwritten OCR Scenarios

Multimodal large language models (MLLMs) are increasingly used as OCR systems in document and knowledge-processing pipelines, but their ability to faithfully read real handwriting remains underexplored. Existing OCR benchmarks focus largely on printed text or clean single-line inputs, leaving limited coverage of realistic handwritten OCR scenarios such as multilingual handwriting, writer errors, and structurally complex mathematical expressions. We introduce OmniHandwritingOCR, a diagnostic benchmark for evaluating MLLMs and OCR systems on handwritten OCR. It covers handwritten text recognition and handwritten mathematical expression recognition across six subtasks and twelve subsets, totaling 77.57K labeled images from public datasets and newly collected student writings. A key component is a difficulty-stratified multi-line formula corpus designed to test robustness under increasing structural complexity. We evaluate thirteen open- and closed-source systems with five complementary metrics under a unified protocol. Results show that current systems remain far from faithful transcription: performance drops sharply on complex multi-line formulas, model rankings vary across language and formula settings, and several generative models hallucinate plausible but visually unsupported corrections. OmniHandwritingOCR provides a challenging testbed for diagnosing language, content, structural, and visual-grounding failure modes of multimodal models in handwritten OCR scenarios.

cs.CV↗
arXiv2026

SPOT: Sparse Probing and Outcome Calibration for On-Policy Distillation

On-policy distillation (OPD) provides dense teacher supervision on student-generated trajectories, but standard reverse-KL training can assign insufficient probability to other plausible continuations. Teacher entropy alone does not reveal whether uncertainty is concentrated among a few plausible next tokens or dispersed over a long probability tail, nor whether the student already represents those candidates well. Moreover, local teacher probabilities may not predict downstream success. We introduce Sparse Probing and Outcome-calibrated Targets OPD (SPOT), which addresses two coupled decisions, where to probe and what to distill, through an acquisition--exploration--exploitation procedure. During acquisition, a position-level score combines normalized teacher entropy, the probability mass captured by a small top-$k$ candidate set, and student--teacher mismatch to allocate a limited probing budget. During exploration, SPOT evaluates teacher-proposed candidates through verifier-scored student continuations. During exploitation, these outcomes produce a closed-form, KL-regularized target that favors candidates with better downstream outcomes while remaining anchored to the teacher distribution. Extensive experiments across multiple student models and reasoning benchmarks demonstrate the effectiveness of SPOT in improving reasoning performance while balancing solution quality and coverage.

cs.LG↗
arXiv2026

Omni-Photonic Base Station: A Three-Functional-Domain Photonic Architecture for Evolutionary 6G Wireless Infrastructure

6G mobile communications impose immersive communication demands for higher data rates and massive connectivity, while emerging integrated sensing-computing-intelligence scenarios require base stations to concurrently enhance computing capability, guarantee service latency, and realize sensing. We propose the Omni-Photonic Base Station-a progressive evolutionary architecture that introduces photonic technologies into three functional domains, namely baseband processing, fronthaul transmission, and the RF front-end, and obtains system-level gains through cross-domain co-design. By introducing photonics, Omni-PBS harnesses the inherent physical advantages of ultra-broad bandwidth, ultra-low propagation latency, and native parallelism to transcend the aforementioned electronic bottlenecks and fulfill the compound demands of 6G scenarios. The optical baseband computing domain employs photonic accelerators to perform linear computation-intensive tasks, improving the energy efficiency of AI inference by one to two orders of magnitude. The analog optical fronthaul domain replaces digital fronthaul with analog radio-over-fiber , which eliminates the ADCs/DACs and digital intermediate-frequency chips in the remote unit and substantially reduces fronthaul bandwidth requirements. The microwave-photonic RF domain breaks through the bandwidth and frequency limitations of electronic RF front-ends via optical true-time-delay beamforming, programmable photonic filtering, and optical heterodyne frequency conversion. The three domains yield system-level gains beyond single-domain summation through four co-design principles: end-to-end optical-domain continuity, cross-domain co-design, optical computing resource scheduling, and joint optimization of functional splitting.

eess.SP↗
arXiv2026

SkillEvo: Self-Renewing Evolution Gradients from Multi-Turn Interaction Feedback

Agent Skills are today either hand-authored or produced in a single LLM generation pass, and consequently possess no closed loop through which they might improve from the interaction failures they actually cause. Recent work does close this loop, but derives its feedback from single-turn question-answering evaluation. The consequence is a sharp asymmetry: once the first round has patched the gaps that a single exchange can reveal, the evolution gradient decays, the defects that surface only across multiple turns remain invisible, and evolution stalls. Governance in these systems is likewise driven by an end-to-end verification score, a scalar gate that can reject a degraded candidate but can neither localize nor repair its structural cause. We argue that the binding constraint on sustained skill evolution is neither editing capability nor the number of iterations, but whether the evaluation feedback keeps supplying trustworthy evolution gradients. We introduce SkillEvo, in which trustworthy feedback generates the gradient and controllable governance constrains its direction. The first component recasts multi-turn user simulation from an evaluation endpoint into a feedback generator: follow-up questions expose defects layer by layer, so that every round of revision both consumes feedback and produces new feedback. The second replaces the passive rejection of a scalar gate with an independent governance layer that actively repairs factual degradation and structural bloat, preventing the gradient from drifting as degradation accumulates. Across six categories of cloud services, 9 production Skills, and 98 skill-reference files, SkillEvo surpasses self-reflection-based evolution by 23.0 points and single- turn-QA-driven evolution by 15.4 points.

cs.AI↗
arXiv2026

COBRA-DOSE: Copula-based Bayesian Model Averaging for Dose Selection

Early-phase clinical trials for dose selection typically enrol few patients and aim to identify doses that are both safe and promising for further study. While traditional approaches identify the maximum tolerated dose, modern trials for targeted therapies often seek the optimal biological dose, defined as the lowest dose achieving sufficient biological activity with acceptable safety. In immunology settings, assessment of biological activity is based on multiple biomarkers or clinical endpoints. Clinicians leading dose-selection efforts would thus benefit from transparent summaries of the probabilities of observing combinations of biomarker outcomes across doses. However, such inference is challenging in small samples where complex modelling assumptions are difficult to verify. To address this limitation, we propose COBRA-DOSE, a framework for posterior predictive inference based on two endpoints that models dependence via copulas and accounts for uncertainty in both marginal distributions and dependence structures through Bayesian model averaging. This approach avoids reliance on a single model and yields interpretable quantities for clinical decision making. We demonstrate the performance of COBRA-DOSE using DEN-181, a phase I immunology trial in rheumatoid arthritis. We also provide a general implementation of our approach through the CobraDose package in R.

stat.ME↗
arXiv2026

Hierarchical Acoustic-Semantic Modeling: Modality Separation and Semantic Coherence for Full-Duplex SLMs

Developing seamless, high-performance, native intelligent full-duplex Spoken Language Models (SLMs) remains a critical challenge and long-standing goal for the speech and NLP community. Despite notable progress, recent endeavors are fundamentally constrained by severe modality interference, which causes substantial knowledge degradation and compromises semantic integrity -- ultimately making full-duplex SLMs feel unnatural and unintelligent. In this paper, through an exhaustive fine-grained analysis of model optimization dynamics, we uncover the root cause of such performance degradation, revealing that modality interference arises from inherent gradient conflicts between acoustic and semantic modeling when the two modalities are forced to share a deep parameter space. Guided by this key insight, we introduce Lychee-FD, a native end-to-end full-duplex framework designed to mitigate modality interference. Importantly, we propose a hierarchical parameter separation strategy that decouples conflicting modalities in deep layers while preserving cross-modality coherence via a dedicated semantic alignment channel. Extensive experiments on multiple full-duplex benchmarks demonstrate that our method significantly advances the state of the art, yielding substantial improvements in both speech intelligence (+7.4% on Spoken QA) and full-duplex interaction fluidity (+28.5% on FullDuplexBench 1.5) without compromising inference efficiency. To the best of our knowledge, this work is the first to achieve two key advances: 1) uncovering and elucidating the root cause of modality interference in full-duplex SLMs, and 2) designing an elegant hierarchical model together with a practical solution for seamless, high-performance, native intelligent full-duplex SLMs.

cs.CL↗