arXiv subjects

Shuo Yang

Publications and source records attributed to Shuo Yang.

At least 217 records · Page 12Linked to original sources

arXiv2025

Sparse VideoGen: Accelerating Video Diffusion Transformers with Spatial-Temporal Sparsity

Diffusion Transformers (DiTs) dominate video generation but their high computational cost severely limits real-world applicability, usually requiring tens of minutes to generate a few seconds of video even on high-performance GPUs. This inefficiency primarily arises from the quadratic computational complexity of 3D Full Attention with respect to the context length. In this paper, we propose a training-free framework termed Sparse VideoGen (SVG) that leverages the inherent sparsity in 3D Full Attention to boost inference efficiency. We reveal that the attention heads can be dynamically classified into two groups depending on distinct sparse patterns: (1) Spatial Head, where only spatially-related tokens within each frame dominate the attention output, and (2) Temporal Head, where only temporally-related tokens across different frames dominate. Based on this insight, SVG proposes an online profiling strategy to capture the dynamic sparse patterns and predicts the type of attention head. Combined with a novel hardware-efficient tensor layout transformation and customized kernel implementations, SVG achieves up to 2.28x and 2.33x end-to-end speedup on CogVideoX-v1.5 and HunyuanVideo, respectively, while preserving generation quality. Our code is open-sourced and is available at https://github.com/svg-project/Sparse-VideoGen

cs.CV↗

arXiv2025

Dynamic Rank Adjustment in Diffusion Policies for Efficient and Flexible Training

Diffusion policies trained via offline behavioral cloning have recently gained traction in robotic motion generation. While effective, these policies typically require a large number of trainable parameters. This model size affords powerful representations but also incurs high computational cost during training. Ideally, it would be beneficial to dynamically adjust the trainable portion as needed, balancing representational power with computational efficiency. For example, while overparameterization enables diffusion policies to capture complex robotic behaviors via offline behavioral cloning, the increased computational demand makes online interactive imitation learning impractical due to longer training time. To address this challenge, we present a framework, called DRIFT, that uses the Singular Value Decomposition to enable dynamic rank adjustment during diffusion policy training. We implement and demonstrate the benefits of this framework in DRIFT-DAgger, an imitation learning algorithm that can seamlessly slide between an offline bootstrapping phase and an online interactive phase. We perform extensive experiments to better understand the proposed framework, and demonstrate that DRIFT-DAgger achieves improved sample efficiency and faster training with minimal impact on model performance. The project website is available at: https://apollo-lab-yale.github.io/25-RSS-DRIFT-website/.

cs.RO↗

arXiv2025

Locally Purified Density Operators for Symmetry-Protected Topological Phases in Mixed States

We propose a tensor network approach known as the locally purified density operator (LPDO) to investigate the classification and characterization of symmetry-protected topological (SPT) phases in open quantum systems. We extend the concept of injectivity, originally associated with matrix product states and projected entangled pair states, to LPDOs in $(1+1)D$ and $(2+1)D$ systems, unveiling two distinct types of injectivity conditions inherent in short-range entangled density matrices. Within the LPDO framework, we outline a classification scheme for decohered average symmetry-protected topological (ASPT) phases, consistent with earlier results obtained through spectrum sequence techniques. We first illustrate our framework with ASPT phases protected by fermion parity symmetry, then extend the classification of ASPT phases to a general group extension. We demonstrate examples of explicit construction of fixed-point LPDOs for ASPT phases including intrinsic ASPTs in both $(1+1)D$ and $(2+1)D$ systems.

cond-mat.str-el↗

arXiv2025

Searching for heavy vector-like B quark via pair production in fully hadronic channels at the CLIC

Vector-like quarks (VLQs) are introduced in many new physics senarios beyond the Standard Model (SM) to address some problems faced by SM. In this paper, we explore the pair production of TeV-scale vector-like B quark (VLQ-$B$) at the future 3 TeV Compact Linear Collider (CLIC) in simplified effective lagrangian framework. We consider the decay modes of $B\rightarrow bZ$ and $B\rightarrow bh$ followed by hadronic decay of $Z$ and $h$ bosons. The large mass of VLQ-$B$ will induce highly boosted bosons $Z$ or $h$ which are more likely to form as fat-jets. By performing a rapid detector simulation of the signal and background events and clustering the jets with a large radius R, signal-background analyses are carried out. And the exclusion limit at the 95\% confidence level and the 5$σ$ discovery prospects are obtained with an integrated luminosity of 5$\text{ab}^{-1}$.

hep-ph↗

arXiv2025

LAMD: Context-driven Android Malware Detection and Classification with LLMs

The rapid growth of mobile applications has escalated Android malware threats. Although there are numerous detection methods, they often struggle with evolving attacks, dataset biases, and limited explainability. Large Language Models (LLMs) offer a promising alternative with their zero-shot inference and reasoning capabilities. However, applying LLMs to Android malware detection presents two key challenges: (1)the extensive support code in Android applications, often spanning thousands of classes, exceeds LLMs' context limits and obscures malicious behavior within benign functionality; (2)the structural complexity and interdependencies of Android applications surpass LLMs' sequence-based reasoning, fragmenting code analysis and hindering malicious intent inference. To address these challenges, we propose LAMD, a practical context-driven framework to enable LLM-based Android malware detection. LAMD integrates key context extraction to isolate security-critical code regions and construct program structures, then applies tier-wise code reasoning to analyze application behavior progressively, from low-level instructions to high-level semantics, providing final prediction and explanation. A well-designed factual consistency verification mechanism is equipped to mitigate LLM hallucinations from the first tier. Evaluation in real-world settings demonstrates LAMD's effectiveness over conventional detectors, establishing a feasible basis for LLM-driven malware analysis in dynamic threat landscapes.

cs.CR↗

arXiv2025

End-to-end Open-vocabulary Video Visual Relationship Detection using Multi-modal Prompting

Open-vocabulary video visual relationship detection aims to expand video visual relationship detection beyond annotated categories by detecting unseen relationships between both seen and unseen objects in videos. Existing methods usually use trajectory detectors trained on closed datasets to detect object trajectories, and then feed these trajectories into large-scale pre-trained vision-language models to achieve open-vocabulary classification. Such heavy dependence on the pre-trained trajectory detectors limits their ability to generalize to novel object categories, leading to performance degradation. To address this challenge, we propose to unify object trajectory detection and relationship classification into an end-to-end open-vocabulary framework. Under this framework, we propose a relationship-aware open-vocabulary trajectory detector. It primarily consists of a query-based Transformer decoder, where the visual encoder of CLIP is distilled for frame-wise open-vocabulary object detection, and a trajectory associator. To exploit relationship context during trajectory detection, a relationship query is embedded into the Transformer decoder, and accordingly, an auxiliary relationship loss is designed to enable the decoder to perceive the relationships between objects explicitly. Moreover, we propose an open-vocabulary relationship classifier that leverages the rich semantic knowledge of CLIP to discover novel relationships. To adapt CLIP well to relationship classification, we design a multi-modal prompting method that employs spatio-temporal visual prompting for visual representation and vision-guided language prompting for language input. Extensive experiments on two public datasets, VidVRD and VidOR, demonstrate the effectiveness of our framework. Our framework is also applied to a more difficult cross-dataset scenario to further demonstrate its generalization ability.

cs.CV↗

arXiv2025

Development of a Miniaturized, Automated, and Cost-Effective Device for Enzyme-Linked Immunosorbent Assay

In this work, a miniaturized, automated, and cost-effective ELISA device is designed and implemented, without the utilization of conventional techniques such as pipetting or microfluidic valve technologies. The device has dimensions of 24 cm x 19 cm x 14 cm and weighs <3 Kg. The total hardware cost of the device is estimated to be approximately $1,200, which can be further reduced through optimization during scale-up production. 3D printed disposable parts, including the reagent reservoir disk and the microfluidic connector, have also been developed. IL-6 is used as a model system to demonstrate how the device provides an ELISA measurement. The cost per test is estimated to be less than ten dollars. The compactness, automated operation, along with the cost-effectiveness of this ELISA device, makes it suitable for point-of-care applications in resource-limited regions.

physics.med-ph↗

arXiv2025

Next Patch Prediction for Autoregressive Visual Generation

Autoregressive models, built based on the Next Token Prediction (NTP) paradigm, show great potential in developing a unified framework that integrates both language and vision tasks. Pioneering works introduce NTP to autoregressive visual generation tasks. In this work, we rethink the NTP for autoregressive image generation and extend it to a novel Next Patch Prediction (NPP) paradigm. Our key idea is to group and aggregate image tokens into patch tokens with higher information density. By using patch tokens as a more compact input sequence, the autoregressive model is trained to predict the next patch, significantly reducing computational costs. To further exploit the natural hierarchical structure of image data, we propose a multi-scale coarse-to-fine patch grouping strategy. With this strategy, the training process begins with a large patch size and ends with vanilla NTP where the patch size is 1$\times$1, thus maintaining the original inference process without modifications. Extensive experiments across a diverse range of model sizes demonstrate that NPP could reduce the training cost to around 0.6 times while improving image generation quality by up to 1.0 FID score on the ImageNet 256x256 generation benchmark. Notably, our method retains the original autoregressive model architecture without introducing additional trainable parameters or specifically designing a custom image tokenizer, offering a flexible and plug-and-play solution for enhancing autoregressive visual generation.

cs.CV↗

arXiv2025

UPME: An Unsupervised Peer Review Framework for Multimodal Large Language Model Evaluation

Multimodal Large Language Models (MLLMs) have emerged to tackle the challenges of Visual Question Answering (VQA), sparking a new research focus on conducting objective evaluations of these models. Existing evaluation methods face limitations due to the significant human workload required to design Q&A pairs for visual images, which inherently restricts the scale and scope of evaluations. Although automated MLLM-as-judge approaches attempt to reduce the human workload through automatic evaluations, they often introduce biases. To address these problems, we propose an Unsupervised Peer review MLLM Evaluation framework. It utilizes only image data, allowing models to automatically generate questions and conduct peer review assessments of answers from other models, effectively alleviating the reliance on human workload. Additionally, we introduce the vision-language scoring system to mitigate the bias issues, which focuses on three aspects: (i) response correctness; (ii) visual understanding and reasoning; and (iii) image-text correlation. Experimental results demonstrate that UPME achieves a Pearson correlation of 0.944 with human evaluations on the MMstar dataset and 0.814 on the ScienceQA dataset, indicating that our framework closely aligns with human-designed benchmarks and inherent human preferences.

cs.CV↗

arXiv2025

NumScout: Unveiling Numerical Defects in Smart Contracts using LLM-Pruning Symbolic Execution

In recent years, the Ethereum platform has witnessed a proliferation of smart contracts, accompanied by exponential growth in total value locked (TVL). High-TVL smart contracts often require complex numerical computations, particularly in mathematical financial models used by many decentralized applications (DApps). Improper calculations can introduce numerical defects, posing potential security risks. Existing research primarily focuses on traditional numerical defects like integer overflow, and there is currently a lack of systematic research and effective detection methods targeting new types of numerical defects. In this paper, we identify five new types of numerical defects through the analysis of 1,199 audit reports by utilizing the open card method. Each defect is defined and illustrated with a code example to highlight its features and potential consequences. We also propose NumScout, a symbolic execution-based tool designed to detect these five defects. Specifically, the tool combines information from source code and bytecode, analyzing key operations such as comparisons and transfers, to effectively locate defects and report them based on predefined detection patterns. Furthermore, NumScout uses a large language model (LLM) to prune functions which are unrelated to numerical operations. This step allows symbolic execution to quickly enter the target function and improve runtime speed by 28.4%. We run NumScout on 6,617 real-world contracts and evaluated its performance based on manually labeled results. We find that 1,774 contracts contained at least one of the five defects, and the tool achieved an overall precision of 89.7%.

cs.SE↗

arXiv2025

Advancing sustainable energy solutions with microfluidic porous media

The transition to a sustainable, low-carbon energy future requires transformative advancements in energy and environmental technologies. Carbon capture and sequestration, underground hydrogen storage, and nuclear waste geological disposal will be central aspects of a sustainable energy future, both for mitigating CO2 emissions and providing green energy. A comprehensive understanding of multiphase flow through porous media, along with reactive transport and microbial activities, is essential for assessing the feasibility and managing the risks of these technologies. Microfluidic porous media platforms have emerged as powerful tools for the direct visualization of multiphase reactive flow in porous media and eventually optimizing these multiple physicochemical and biological processes. This review highlights critical scientific challenges associated with these sustainable energy solutions and summarizes the state-of-the-art microfluidic techniques for studying the interplay between multiphase flow, reactive transport, and biological effects in porous media. We provide a comprehensive overview of how these microfluidic approaches enhance the understanding of fundamental pore-scale dynamics and bridge the gap between pore-scale events and large-scale processes. This review is expected to promote both experimental and theoretical understanding of multiphase reactive flow in porous media, thereby informing material design, process optimization, and predictive modeling for scalable implementation. By fostering interdisciplinary collaboration across microfluidics, fluid mechanics, geophysics, materials science, and subsurface engineering, we hope to accelerate innovation and advance sustainable energy solutions.

physics.flu-dyn↗

arXiv2025

Towards Edge General Intelligence via Large Language Models: Opportunities and Challenges

Edge Intelligence (EI) has been instrumental in delivering real-time, localized services by leveraging the computational capabilities of edge networks. The integration of Large Language Models (LLMs) empowers EI to evolve into the next stage: Edge General Intelligence (EGI), enabling more adaptive and versatile applications that require advanced understanding and reasoning capabilities. However, systematic exploration in this area remains insufficient. This survey delineates the distinctions between EGI and traditional EI, categorizing LLM-empowered EGI into three conceptual systems: centralized, hybrid, and decentralized. For each system, we detail the framework designs and review existing implementations. Furthermore, we evaluate the performance and throughput of various Small Language Models (SLMs) that are more suitable for development on edge devices. This survey provides researchers with a comprehensive vision of EGI, offering insights into its vast potential and establishing a foundation for future advancements in this rapidly evolving field.

cs.DC↗

arXiv2025

WorldModelBench: Judging Video Generation Models As World Models

Video generation models have rapidly progressed, positioning themselves as video world models capable of supporting decision-making applications like robotics and autonomous driving. However, current benchmarks fail to rigorously evaluate these claims, focusing only on general video quality, ignoring important factors to world models such as physics adherence. To bridge this gap, we propose WorldModelBench, a benchmark designed to evaluate the world modeling capabilities of video generation models in application-driven domains. WorldModelBench offers two key advantages: (1) Against to nuanced world modeling violations: By incorporating instruction-following and physics-adherence dimensions, WorldModelBench detects subtle violations, such as irregular changes in object size that breach the mass conservation law - issues overlooked by prior benchmarks. (2) Aligned with large-scale human preferences: We crowd-source 67K human labels to accurately measure 14 frontier models. Using our high-quality human labels, we further fine-tune an accurate judger to automate the evaluation procedure, achieving 8.6% higher average accuracy in predicting world modeling violations than GPT-4o with 2B parameters. In addition, we demonstrate that training to align human annotations by maximizing the rewards from the judger noticeably improve the world modeling capability. The website is available at https://worldmodelbench-team.github.io.

cs.CV↗

arXiv2025

WakeMint: Detecting Sleepminting Vulnerabilities in NFT Smart Contracts

The non-fungible tokens (NFTs) market has evolved over the past decade, with NFTs serving as unique digital identifiers on a blockchain that certify ownership and authenticity. However, their high value also attracts attackers who exploit vulnerabilities in NFT smart contracts for illegal profits, thereby harming the NFT ecosystem. One notable vulnerability in NFT smart contracts is sleepminting, which allows attackers to illegally transfer others' tokens. Although some research has been conducted on sleepminting, these studies are basically qualitative analyses or based on historical transaction data. There is a lack of understanding from the contract code perspective, which is crucial for identifying such issues and preventing attacks before they occur. To address this gap, in this paper, we categoriz four distinct types of sleepminting in NFT smart contracts. Each type is accompanied by a comprehensive definition and illustrative code examples to provide how these vulnerabilities manifest within the contract code. Furthermore, to help detect the defined defects before the sleepminting problem occurrence, we propose a tool named WakeMint, which is built on a symbolic execution framework and is designed to be compatible with both high and low versions of Solidity. The tool also employs a pruning strategy to shorten the detection period. Additionally, WakeMint gathers some key information, such as the owner of an NFT and emissions of events related to the transfer of the NFT's ownership during symbolic execution. Then, it analyzes the features of the transfer function based on this information so that it can judge the existence of sleepminting. We ran WakeMint on 11,161 real-world NFT smart contracts and evaluated the results. We found 115 instances of sleepminting issues in total, and the precision of our tool is 87.8%.

cs.SE↗

arXiv2025

P-TA: Using Proximal Policy Optimization to Enhance Tabular Data Augmentation via Large Language Models

A multitude of industries depend on accurate and reasonable tabular data augmentation for their business processes. Contemporary methodologies in generating tabular data revolve around utilizing Generative Adversarial Networks (GAN) or fine-tuning Large Language Models (LLM). However, GAN-based approaches are documented to produce samples with common-sense errors attributed to the absence of external knowledge. On the other hand, LLM-based methods exhibit a limited capacity to capture the disparities between synthesized and actual data distribution due to the absence of feedback from a discriminator during training. Furthermore, the decoding of LLM-based generation introduces gradient breakpoints, impeding the backpropagation of loss from a discriminator, thereby complicating the integration of these two approaches. To solve this challenge, we propose using proximal policy optimization (PPO) to apply GANs, guiding LLMs to enhance the probability distribution of tabular features. This approach enables the utilization of LLMs as generators for GANs in synthesizing tabular data. Our experiments demonstrate that PPO leads to an approximately 4\% improvement in the accuracy of models trained on synthetically generated data over state-of-the-art across three real-world datasets.

cs.LG↗

arXiv2025

PiCO: Peer Review in LLMs based on the Consistency Optimization

Existing large language models (LLMs) evaluation methods typically focus on testing the performance on some closed-environment and domain-specific benchmarks with human annotations. In this paper, we explore a novel unsupervised evaluation direction, utilizing peer-review mechanisms to measure LLMs automatically. In this setting, both open-source and closed-source LLMs lie in the same environment, capable of answering unlabeled questions and evaluating each other, where each LLM's response score is jointly determined by other anonymous ones. To obtain the ability hierarchy among these models, we assign each LLM a learnable capability parameter to adjust the final ranking. We formalize it as a constrained optimization problem, intending to maximize the consistency of each LLM's capabilities and scores. The key assumption behind is that high-level LLM can evaluate others' answers more accurately than low-level ones, while higher-level LLM can also achieve higher response scores. Moreover, we propose three metrics called PEN, CIN, and LIS to evaluate the gap in aligning human rankings. We perform experiments on multiple datasets with these metrics, validating the effectiveness of the proposed approach.

cs.CL↗

arXiv2025

Revisiting the Index Construction of Proximity Graph-Based Approximate Nearest Neighbor Search

Proximity graphs (PG) have gained increasing popularity as the state-of-the-art solutions to $k$-approximate nearest neighbor ($k$-ANN) search on high-dimensional data, which serves as a fundamental function in various fields, e.g., retrieval-augmented generation. Although PG-based approaches have the best $k$-ANN search performance, their index construction cost is superlinear to the number of points. Such superlinear cost substantially limits their scalability in the era of big data. Hence, the goal of this paper is to accelerate the construction of PG-based methods without compromising their $k$-ANN search performance. To achieve this goal, two mainstream categories of PG are revisited: relative neighborhood graph (RNG) and navigable small world graph (NSWG). By revisiting their construction process, we find the issues of construction efficiency. To address these issues, we propose a new construction framework with a novel pruning strategy for edge selection, which accelerates RNG construction while keeping its $k$-ANN search performance. Then, we integrate this framework into NSWG construction to enhance both the construction efficiency and $k$-ANN search performance of NSWG. Extensive experiments are conducted to validate our construction framework for both RNG and NSWG, and that it significantly reduces the PG construction cost, achieving up to 5.6x speedup, while not compromising the $k$-ANN search performance.

cs.DB↗

arXiv2025

Continual Adaptation for Autonomous Driving with the Mixture of Progressive Experts Network

Learning-based autonomous driving requires continuous integration of diverse knowledge in complex traffic , yet existing methods exhibit significant limitations in adaptive capabilities. Addressing this gap demands autonomous driving systems that enable continual adaptation through dynamic adjustments to evolving environmental interactions. This underscores the necessity for enhanced continual learning capabilities to improve system adaptability. To address these challenges, the paper introduces a dynamic progressive optimization framework that facilitates adaptation to variations in dynamic environments, achieved by integrating reinforcement learning and supervised learning for data aggregation. Building on this framework, we propose the Mixture of Progressive Experts (MoPE) network. The proposed method selectively activates multiple expert models based on the distinct characteristics of each task and progressively refines the network architecture to facilitate adaptation to new tasks. Simulation results show that the MoPE model outperforms behavior cloning methods, achieving up to a 7.8% performance improvement in intricate urban road environments.

cs.RO↗