arXiv subjects

Miao Zhang

Publications and source records attributed to Miao Zhang.

At least 73 records · Page 4Linked to original sources

arXiv2026

CVeDRL: An Efficient Code Verifier via Difficulty-aware Reinforcement Learning

Code verifiers play a critical role in post-verification for LLM-based code generation, yet existing supervised fine-tuning methods suffer from data scarcity, high failure rates, and poor inference efficiency. While reinforcement learning (RL) offers a promising alternative by optimizing models through execution-driven rewards without labeled supervision, our preliminary results show that naive RL with only functionality rewards fails to generate effective unit tests for difficult branches and samples. We first theoretically analyze showing that branch coverage, sample difficulty, syntactic and functional correctness can be jointly modeled as RL rewards, where optimizing these signals can improve the reliability of unit-test-based verification. Guided by this analysis, we design syntax- and functionality-aware rewards and further propose branch- and sample-difficulty--aware RL using exponential reward shaping and static analysis metrics. With this formulation, CVeDRL achieves state-of-the-art performance with only 0.6B parameters, yielding up to 28.97% higher pass rate and 15.08% higher branch coverage than GPT-3.5, while delivering over $20\times$ faster inference than competitive baselines. Code is available at https://github.com/LIGHTCHASER1/CVeDRL.git

cs.AI↗

arXiv2026

ViTMAlis: Towards Latency-Critical Mobile Video Analytics with Vision Transformers

Edge-assisted mobile video analytics (MVA) applications are increasingly shifting from using vision models based on convolutional neural networks (CNNs) to those built on vision transformers (ViTs) to leverage their superior global context modeling and generalization capabilities. However, deploying these advanced models in latency-critical MVA scenarios presents significant challenges. Unlike traditional CNN-based offloading paradigms where network transmission is the primary bottleneck, ViT-based systems are constrained by substantial inference delays, particularly for dense prediction tasks where the need for high-resolution inputs exacerbates the inherent quadratic computational complexity of ViTs. To address these challenges, we propose a dynamic mixed-resolution inference strategy tailored for ViT-backboned dense prediction models, enabling flexible runtime trade-offs between speed and accuracy. Building on this, we introduce ViTMAlis, a ViT-native device-to-edge offloading framework that dynamically adapts to network conditions and video content to jointly reduce transmission and inference delays. We implement a fully functional prototype of ViTMAlis on commodity mobile and edge devices. Extensive experiments demonstrate that, compared to state-of-the-art accuracy-centric, content-aware, and latency-adaptive baselines, ViTMAlis significantly reduces end-to-end offloading latency while improving user-perceived rendering accuracy, providing a practical foundation for next-generation mobile intelligence.

cs.NI↗

arXiv2026

Dynamic Role Assignment for Multi-Agent Debate

Multi-agent large language model (LLM) and vision-language model (VLM) debate systems employ specialized roles for complex problem-solving, yet model specializations are not leveraged to decide which model should fill which role. We propose dynamic role assignment, a framework that runs a Meta-Debate to select suitable agents before the actual debate. The meta-debate has two stages: (1) proposal, where candidates provide role-tailored arguments, and (2) peer review, where proposals are scored with data and role-specific criteria to choose the best agent for each position. We evaluate our method on LLM problem solving benchmarks. Applied on top of existing debate systems, our approach consistently outperforms uniform assignments (filling all roles with the same model) by up to 74.8% and random assignments (assigning models to roles without considering their suitability) by up to 29.7%, depending on the task and the specific assignment. This work establishes a new paradigm for multi-agent system design, shifting from static agent deployment to dynamic and capability-aware selection.

cs.CL↗

arXiv2026

MyGram: Modality-aware Graph Transformer with Global Distribution for Multi-modal Entity Alignment

Multi-modal entity alignment aims to identify equivalent entities between two multi-modal Knowledge graphs by integrating multi-modal data, such as images and text, to enrich the semantic representations of entities. However, existing methods may overlook the structural contextual information within each modality, making them vulnerable to interference from shallow features. To address these challenges, we propose MyGram, a modality-aware graph transformer with global distribution for multi-modal entity alignment. Specifically, we develop a modality diffusion learning module to capture deep structural contextual information within modalities and enable fine-grained multi-modal fusion. In addition, we introduce a Gram Loss that acts as a regularization constraint by minimizing the volume of a 4-dimensional parallelotope formed by multi-modal features, thereby achieving global distribution consistency across modalities. We conduct experiments on five public datasets. Results show that MyGram outperforms baseline models, achieving a maximum improvement of 4.8% in Hits@1 on FBDB15K, 9.9% on FBYG15K, and 4.3% on DBP15K.

cs.AI↗

arXiv2026

Identity-Robust Language Model Generation via Content Integrity Preservation

Large Language Model (LLM) outputs often vary across user sociodemographic attributes, leading to disparities in factual accuracy, utility, and safety, even for objective questions where demographic information is irrelevant. Unlike prior work on stereotypical or representational bias, this paper studies identity-dependent degradation of core response quality. We show empirically that such degradation arises from biased generation behavior, despite factual knowledge being robustly encoded across identities. Motivated by this mismatch, we propose a lightweight, training-free framework for identity-robust generation that selectively neutralizes non-critical identity information while preserving semantically essential attributes, thus maintaining output content integrity. Experiments across four benchmarks and 18 sociodemographic identities demonstrate an average 77% reduction in identity-dependent bias compared to vanilla prompting and a 45% reduction relative to prompt-based defenses. Our work addresses a critical gap in mitigating the impact of user identity cues in prompts on core generation quality.

cs.CL↗

arXiv2026

Earthquake Source Depth Determination using Single Station Waveforms and Deep Learning

In areas with limited station coverage, earthquake depth constraints are much less accurate than their latitude and longitude. Traditional travel-time-based location methods struggle to constrain depths due to imperfect station distribution and the strong trade-off between source depth and origin time. Identifying depth phases at regional distances is usually hindered by strong wave scattering, which is particularly challenging for low-magnitude events. Deep learning algorithms, capable of extracting various features from seismic waveforms, including phase arrivals, phase amplitudes, as well as phase frequency, offer promising constraints to earthquake depths. In this work, we propose a novel depth feature extraction network (named VGGDepth), which directly maps seismic waveforms to earthquake depth using three-component waveforms. The network structure is adapted from VGG16 in computer vision. It is designed to take single-station three-component waveforms as inputs and produce depths as outputs, achieving a direct mapping from waveforms to depths. Two scenarios are considered in our model development: (1) training and testing solely on the same seismic station, and (2) generalizing by training and testing on different seismic stations within a particular region. We demonstrate the efficacy of our methodology using seismic data from the 2016-2017 Central Apennines, Italy earthquake sequence. Results demonstrate that earthquake depths can be estimated from single stations with uncertainties of hundreds of meters. These uncertainties are further reduced by averaging results from multiple stations. Our method shows strong potential for earthquake depth determination, particularly for events recorded by single or sparsely distributed stations, such as historically instrumented earthquakes.

physics.geo-ph↗

arXiv2026

TBPLaS 2.0: a Tight-Binding Package for Large-scale Simulation

The common exact diagonalization-based techniques to solving tight-binding models suffer from O(N^2) and O(N^3) scaling with respect to model size in memory and CPU time, hindering their applications in large tight-binding models. On the contrary, the tight-binding propagation method (TBPM) can achieve linear scaling in both memory and CPU time, and is capable of handling large tight-binding models with billions of orbitals. In this paper, we introduce version 2.0 of TBPLaS, a package for large-scale simulation based on TBPM. This new version brings significant improvements with many new features. Existing Python/Cython modeling tools have been thoroughly optimized, and a compatible C++ implementation of the modeling tools is now available, offering efficiency enhancement of several orders. The solvers have been rewritten in C++ from scratch, with the efficiency enhanced by several times or even by an order of magnitude. The workflow of utilizing solvers has also been unified into a more comprehensive and consistent manner. New features include spin texture, Berry curvature and Chern number calculation, partial diagonalization for specific eigenvalues and eigenstates, analytical Hamiltonian, and GPU computing support. The documentation and tutorials have also been updated to the new version. In this paper, we discuss the revisions with respect to version 1.3 and demonstrate the new features. Benchmarks on modeling tools and solvers are also provided.

physics.comp-ph↗

arXiv2026

MacVQA: Adaptive Memory Allocation and Global Noise Filtering for Continual Visual Question Answering

Visual Question Answering (VQA) requires models to reason over multimodal information, combining visual and textual data. With the development of continual learning, significant progress has been made in retaining knowledge and adapting to new information in the VQA domain. However, current methods often struggle with balancing knowledge retention, adaptation, and robust feature representation. To address these challenges, we propose a novel framework with adaptive memory allocation and global noise filtering called MacVQA for visual question answering. MacVQA fuses visual and question information while filtering noise to ensure robust representations, and employs prototype-based memory allocation to optimize feature quality and memory usage. These designs enable MacVQA to balance knowledge acquisition, retention, and compositional generalization in continual VQA learning. Experiments on ten continual VQA tasks show that MacVQA outperforms existing baselines, achieving 43.38% average accuracy and 2.32% average forgetting on standard tasks, and 42.53% average accuracy and 3.60% average forgetting on novel composition tasks.

cs.CV↗

arXiv2025

KeenKT: Knowledge Mastery-State Disambiguation for Knowledge Tracing

Knowledge Tracing (KT) aims to dynamically model a student's mastery of knowledge concepts based on their historical learning interactions. Most current methods rely on single-point estimates, which cannot distinguish true ability from outburst or carelessness, creating ambiguity in judging mastery. To address this issue, we propose a Knowledge Mastery-State Disambiguation for Knowledge Tracing model (KeenKT), which represents a student's knowledge state at each interaction using a Normal-Inverse-Gaussian (NIG) distribution, thereby capturing the fluctuations in student learning behaviors. Furthermore, we design an NIG-distance-based attention mechanism to model the dynamic evolution of the knowledge state. In addition, we introduce a diffusion-based denoising reconstruction loss and a distributional contrastive learning loss to enhance the model's robustness. Extensive experiments on six public datasets demonstrate that KeenKT outperforms SOTA KT models in terms of prediction accuracy and sensitivity to behavioral fluctuations. The proposed method yields the maximum AUC improvement of 5.85% and the maximum ACC improvement of 6.89%.

cs.AI↗

arXiv2025

FuncGenFoil: Airfoil Generation and Editing Model in Function Space

Aircraft manufacturing is the jewel in the crown of industry, in which generating high-fidelity airfoil geometries with controllable and editable representations remains a fundamental challenge. Existing deep learning methods, which typically rely on predefined parametric representations (e.g., Bézier) or discrete point sets, face an inherent trade-off between expressive power and resolution adaptability. To tackle this challenge, we introduce FuncGenFoil, a novel function-space generative model that directly reconstructs airfoil geometries as function curves. Our method inherits the advantages of arbitrary-resolution sampling and smoothness from parametric functions, as well as the strong expressiveness of discrete point-based representations. Empirical evaluations demonstrate that FuncGenFoil improves upon state-of-the-art methods in airfoil generation, achieving a relative 74.4% reduction in label error and a 23.2% increase in diversity on the AF-200K dataset. Our results highlight the advantages of function-space modeling for aerodynamic shape optimization, offering a powerful and flexible framework for high-fidelity airfoil design.

cs.LG↗

arXiv2025

DOS: Distilling Observable Softmaps of Zipfian Prototypes for Self-Supervised Point Representation

Recent advances in self-supervised learning (SSL) have shown tremendous potential for learning 3D point cloud representations without human annotations. However, SSL for 3D point clouds still faces critical challenges due to irregular geometry, shortcut-prone reconstruction, and unbalanced semantics distribution. In this work, we propose DOS (Distilling Observable Softmaps), a novel SSL framework that self-distills semantic relevance softmaps only at observable (unmasked) points. This strategy prevents information leakage from masked regions and provides richer supervision than discrete token-to-prototype assignments. To address the challenge of unbalanced semantics in an unsupervised setting, we introduce Zipfian prototypes and incorporate them using a modified Sinkhorn-Knopp algorithm, Zipf-Sinkhorn, which enforces a power-law prior over prototype usage and modulates the sharpness of the target softmap during training. DOS outperforms current state-of-the-art methods on semantic segmentation and 3D object detection across multiple benchmarks, including nuScenes, Waymo, SemanticKITTI, ScanNet, and ScanNet200, without relying on extra data or annotations. Our results demonstrate that observable-point softmaps distillation offers a scalable and effective paradigm for learning robust 3D representations.

cs.CV↗

arXiv2025

Configurable Fairness: Direct Optimization of Parity Metrics via Vision-Language Models

Performance disparities of image recognition across demographic groups are known to exist in deep learning-based models, due to imbalanced group representations or spurious correlation between group and target labels. Previous work has addressed such challenges without relying on expensive group labels, typically by upweighting high-loss samples or balancing discovered clusters. However, these heuristic strategies lack direct connection to specific fairness metrics and cannot guarantee optimization of parity-based criteria like equal opportunity, which ensures equal chance to receive positive outcomes across groups. In this work, we propose a novel paradigm that directly optimizes parity-based fairness metrics through specifically designed training objectives, without requiring group labels. We leverage vision-language models to analyze sensitive attribute relevancy for individual samples, then formulate loss functions that mathematically connect to each target fairness metric. This enables flexible optimization of different fairness criteria based on application needs. Experiments on multiple image classification datasets show that our metric-specific approach significantly improves parity-based fairness criteria and outperforms existing methods.

cs.CV↗

arXiv2025

Elasto-plastic cell-based smoothed finite element method solving geotechnical problems

This work develops an elasto-plastic cell-based smoothed finite element method (CSFEM) for geotechnical analysis. The formulation incorporates a smoothed strain field into the standard elasto-plastic framework based on the Mohr-Coulomb criterion and is implemented in ABAQUS through a user-defined element (UEL). A UEL-UMAT data-transfer strategy is introduced to enable post-processing of stress and strain in ABAQUS. The method is assessed using several benchmark problems, including three classical examples, a tunnel excavation, and a slope stability analysis. The results show that the CSFEM achieves accuracy comparable to or slightly better than the conventional FEM and matches analytical or reference solutions for the examined cases. These findings indicate that the proposed CSFEM provides a practical and robust alternative for routine elasto-plastic analyses in geotechnical engineering.

math.NA↗

arXiv2025

Class-Aware PillarMix: Can Mixed Sample Data Augmentation Enhance 3D Object Detection with Radar Point Clouds?

Due to the significant effort required for data collection and annotation in 3D perception tasks, mixed sample data augmentation (MSDA) has been widely studied to generate diverse training samples by mixing existing data. Recently, many MSDA techniques have been developed for point clouds, but they mainly target LiDAR data, leaving their application to radar point clouds largely unexplored. In this paper, we examine the feasibility of applying existing MSDA methods to radar point clouds and identify several challenges in adapting these techniques. These obstacles stem from the radar's irregular angular distribution, deviations from a single-sensor polar layout in multi-radar setups, and point sparsity. To address these issues, we propose Class-Aware PillarMix (CAPMix), a novel MSDA approach that applies MixUp at the pillar level in 3D point clouds, guided by class labels. Unlike methods that rely a single mix ratio to the entire sample, CAPMix assigns an independent ratio to each pillar, boosting sample diversity. To account for the density of different classes, we use class-specific distributions: for dense objects (e.g., large vehicles), we skew ratios to favor points from another sample, while for sparse objects (e.g., pedestrians), we sample more points from the original. This class-aware mixing retains critical details and enriches each sample with new information, ultimately generating more diverse training data. Experimental results demonstrate that our method not only significantly boosts performance but also outperforms existing MSDA approaches across two datasets (Bosch Street and K-Radar). We believe that this straightforward yet effective approach will spark further investigation into MSDA techniques for radar data.

cs.CV↗

arXiv2025

Learning from N-Tuple Data with M Positive Instances: Unbiased Risk Estimation and Theoretical Guarantees

Weakly supervised learning often operates with coarse aggregate signals rather than instance labels. We study a setting where each training example is an $n$-tuple containing exactly m positives, while only the count m per tuple is observed. This NTMP (N-tuple with M positives) supervision arises in, e.g., image classification with region proposals and multi-instance measurements. We show that tuple counts admit a trainable unbiased risk estimator (URE) by linking the tuple-generation process to latent instance marginals. Starting from fixed (n,m), we derive a closed-form URE and extend it to variable tuple sizes, variable counts, and their combination. Identification holds whenever the effective mixing rate is separated from the class prior. We establish generalization bounds via Rademacher complexity and prove statistical consistency with standard rates under mild regularity assumptions. To improve finite-sample stability, we introduce simple ReLU corrections to the URE that preserve asymptotic correctness. Across benchmarks converted to NTMP tasks, the approach consistently outperforms representative weak-supervision baselines and yields favorable precision-recall and F1 trade-offs. It remains robust under class-prior imbalance and across diverse tuple configurations, demonstrating that count-only supervision can be exploited effectively through a theoretically grounded and practically stable objective.

cs.LG↗

arXiv2025

Structurally Refined Graph Transformer for Multimodal Recommendation

Multimodal recommendation systems utilize various types of information, including images and text, to enhance the effectiveness of recommendations. The key challenge is predicting user purchasing behavior from the available data. Current recommendation models prioritize extracting multimodal information while neglecting the distinction between redundant and valuable data. They also rely heavily on a single semantic framework (e.g., local or global semantics), resulting in an incomplete or biased representation of user preferences, particularly those less expressed in prior interactions. Furthermore, these approaches fail to capture the complex interactions between users and items, limiting the model's ability to meet diverse users. To address these challenges, we present SRGFormer, a structurally optimized multimodal recommendation model. By modifying the transformer for better integration into our model, we capture the overall behavior patterns of users. Then, we enhance structural information by embedding multimodal information into a hypergraph structure to aid in learning the local structures between users and items. Meanwhile, applying self-supervised tasks to user-item collaborative signals enhances the integration of multimodal information, thereby revealing the representational features inherent to the data's modality. Extensive experiments on three public datasets reveal that SRGFormer surpasses previous benchmark models, achieving an average performance improvement of 4.47 percent on the Sports dataset. The code is publicly available online.

cs.IR↗

arXiv2025

Cost-Sensitive Unbiased Risk Estimation for Multi-Class Positive-Unlabeled Learning

Positive--Unlabeled (PU) learning considers settings in which only positive and unlabeled data are available, while negatives are missing or left unlabeled. This situation is common in real applications where annotating reliable negatives is difficult or costly. Despite substantial progress in PU learning, the multi-class case (MPU) remains challenging: many existing approaches do not ensure \emph{unbiased risk estimation}, which limits performance and stability. We propose a cost-sensitive multi-class PU method based on \emph{adaptive loss weighting}. Within the empirical risk minimization framework, we assign distinct, data-dependent weights to the positive and \emph{inferred-negative} (from the unlabeled mixture) loss components so that the resulting empirical objective is an unbiased estimator of the target risk. We formalize the MPU data-generating process and establish a generalization error bound for the proposed estimator. Extensive experiments on \textbf{eight} public datasets, spanning varying class priors and numbers of classes, show consistent gains over strong baselines in both accuracy and stability.

cs.LG↗

arXiv2025

Demystifying Cookie Sharing Risks in WebView-based Mobile App-in-app Ecosystems

Mini-programs, an emerging mobile application paradigm within super-apps, offer a seamless and installation-free experience. However, the adoption of the web-view component has disrupted their isolation mechanisms, exposing new attack surfaces and vulnerabilities. In this paper, we introduce a novel vulnerability called Cross Mini-program Cookie Sharing (CMCS), which arises from the shared web-view environment across mini-programs. This vulnerability allows unauthorized data exchange across mini-programs by enabling one mini-program to access cookies set by another within the same web-view context, violating isolation principles. As a preliminary step, we analyzed the web-view mechanisms of four major platforms, including WeChat, AliPay, TikTok, and Baidu, and found that all of them are affected by CMCS vulnerabilities. Furthermore, we demonstrate the collusion attack enabled by CMCS, where privileged mini-programs exfiltrate sensitive user data via cookies accessible to unprivileged mini-programs. To measure the impact of collusion attacks enabled by CMCS vulnerabilities in the wild, we developed MiCoScan, a static analysis tool that detects mini-programs affected by CMCS vulnerabilities. MiCoScan employs web-view context modeling to identify clusters of mini-programs sharing the same web-view domain and cross-webview data flow analysis to detect sensitive data transmissions to/from web-views. Using MiCoScan, we conducted a large-scale analysis of 351,483 mini-programs, identifying 45,448 clusters sharing web-view domains, 7,965 instances of privileged data transmission, and 9,877 mini-programs vulnerable to collusion attacks. Our findings highlight the widespread prevalence and significant security risks posed by CMCS vulnerabilities, underscoring the urgent need for improved isolation mechanisms in mini-program ecosystems.

cs.CR↗