arXiv subjects

Yiying Zhang

Publications and source records attributed to Yiying Zhang.

At least 19 recordsLinked to original sources

arXiv2026

Efficient Algorithm Design of Dynamic Spectrum Access by Whittle Index

Dynamic spectrum access problem is an important problem that allows a wireless sub-network to use channels temporarily unoccupied by the parent network for minimizing the spectrum waste. Previous work has shown that the sequential channel allocation problem for the sub-network can be formulated within the restless multi-armed bandits (RMAB) framework. The objective is to maximize the expected long-term return over an infinite horizon while minimizing interference to the parent network. Different from the previous work that exploits a binary feedback (e.g., ACK/NAK) to compensate for sensing errors, we leverage the finer and more robust channel quality indicator (CQI) feedback to update the information state (belief vector) of the sub-network. However, the implementation of CQI-based observation model yields significantly more complex belief transition behaviors in an infinite state space and worsens the curse of dimensionality of dynamic programming. To overcome this challenge, we dive into the rich structures of the value functions and obtain tight bounds on their derivatives. These results lead to the proof of optimality of threshold policies on a single-channel problem with subsidy and subsequently a closed-form channel index function using an iterative method to approximate the well-known Whittle index policy, which offers a low-complexity solution for ranking the currently available channels whose states are never directly observable. Through extensive numerical studies, we demonstrate the superior performance and robustness of our proposed algorithm.

math.OC↗

arXiv2026

OSWorld-Human: Benchmarking the Efficiency of Computer-Use Agents

Generative AI is being leveraged to solve a variety of computer-use tasks involving desktop applications. State-of-the-art systems have focused solely on improving accuracy on leading benchmarks. However, these systems are practically unusable due to extremely high end-to-end latency (e.g., tens of minutes) for tasks that typically take humans just a few minutes to complete. To understand the cause behind this and to guide future developments of computer agents, we conduct the first study on the temporal performance of computer-use agents on OSWorld, the flagship benchmark in computer-use AI. We find that large model calls for planning, reflection, and judging account for most of the overall latency, and as an agent uses more steps to complete a task, each successive step can take 3x longer than steps at the beginning of a task. We then construct OSWorld Human, a manually annotated version of the original OSWorld dataset that contains a human-determined trajectory for each task. We evaluate 16 agents on their efficiency using OSWorld Human and found that even the best agents take 2.7-4.3x more steps than necessary.

cs.AI↗

arXiv2026

TClone: Low-Latency Forking of Live GUI Environments for Computer-Use Agents

Computer-use agents increasingly operate inside live personal workspaces, where their actions can modify files, applications, GUI state, credentials, and authenticated sessions. This creates a tension between safety and quality: agents need isolation and rollback to avoid damaging user state, but also need fast branching to support speculative execution and parallel search. Existing VMs, containers, and checkpoint/restore systems can isolate or recover workloads, but they do not provide low-latency versioning of a full interactive workspace. We present TClone, a forkable personal workspace system for computer-use agents. TClone enables a live GUI workspace to be snapshotted, forked into isolated branches, rolled back, and selectively committed or merged. Its design separates fast branch creation from durable checkpointing, using sibling containers, copy-on-write memory sharing, filesystem versioning, GUI-local execution, and asynchronous checkpointing. In our end-to-end agent-loop measurement, TClone reduces total task latency by 1.9x and 1.5x over KVM and CRIU. By making workspace versioning a first-class systems primitive, TClone supports safer and higher-quality agent execution over real personal computing environments.

cs.OS↗

arXiv2026

GLEAM: A Multimodal Imaging Dataset and HAMM for Glaucoma Classification

We propose glaucoma lesion evaluation and analysis with multimodal imaging (GLEAM), the first publicly available tri-modal glaucoma dataset comprising scanning laser ophthalmoscopy fundus images, circumpapillary OCT images, and visual field pattern deviation maps, annotated with four disease stages, enabling effective exploitation of multimodal complementary information and facilitating accurate diagnosis and treatment across disease stages. To effectively integrate cross-modal information, we propose hierarchical attentive masked modeling (HAMM) for multimodal glaucoma classification. Our framework employs hierarchical attentive encoders and light decoders to focus cross-modal representation learning on the encoder.

eess.IV↗

arXiv2026

Nitsum: Serving Tiered LLM Requests with Adaptive Tensor Parallelism

LLM serving is increasingly multi-tenant: the same deployment must handle latency-critical interactive requests and more relaxed background workloads under a fixed GPU budget. This creates a tiered-SLO setting where maximizing overall goodput (requests that satisfy both TTFT and TPOT targets) is challenging because workload mix, request lengths, and load intensity vary over time. Existing systems mainly optimize request-level controls (e.g., queuing and batching) while keeping execution configuration largely static, which limits adaptation under multi-tier contention. We present Nitsum, a distributed LLM serving system that treats tensor parallelism (TP) as a first-class runtime control surface rather than a static deployment choice. Nitsum jointly optimizes TP level, prefill/decode GPU split, and request scheduling. To make frequent TP adaptation practical, Nitsum introduces TP-aware weight reuse and fast KV migration. Experiments on real traces and targeted microbenchmarks show that Nitsum improves SLO-compliant goodput over SoTA by up to 5.3 times.

cs.DC↗

arXiv2026

VDCores: Resource Decoupled Programming and Execution for Asynchronous GPU

Modern GPUs increasingly rely on specialized and asynchronous hardware units to deliver high performance. Yet these units are often underutilized because today's GPU software stacks still organize programming and execution around a monolithic kernel model that mismatches asynchronous hardware. To address this issue, Virtual Decoupled Engines (VDCores) presents a new decoupled programming and execution model for asynchronous GPUs. VDCores abstracts asynchronous hardware execution units as resource isolated virtual cores and represents workloads as dependency-connected micro-operations (micro-ops). this abstraction removes static orchestration from the programmer, enables automatic overlap of memory and compute based on dependency and resource readiness, and thereby improves utilization of asynchronous hardware resources. Realizing such a decoupled abstraction efficiently on today's GPUs is itself challenging, VDCores addresses this through a GPU-specialized programming model and GPU runtime design that preserves the flexibility while minimizing implementation overhead. Across four LLM inference workloads on GH200, H100, and RTX 6000 Pro GPUs, VDCores significantly improves decoding throughput by 24% on average and by up to 77% under dynamic inputs, while reducing kernel programming and specialization effort by 90%. We have open sourced VDCores at https://github.com/vdcores/vdcores.

cs.DC↗

arXiv2026

Distributionally Robust Insurance under Bregman-Wasserstein Divergence

This paper investigates two optimal insurance contracting problems under distributional uncertainty from the perspective of a potential policyholder, utilizing a Bregman-Wasserstein (BW) ball to characterize the ambiguity set of loss distributions. Unlike the $p$-Wasserstein distance, BW divergence enables asymmetric penalization of deviations from the benchmark distribution. The first problem examines an insurance demand model where the policyholder adopts an $α$-maxmin preference with Value-at-Risk (VaR). We derive the optimal indemnity function in closed form and study, both analytically and numerically, how the asymmetry inherent in BW divergence influences the optimal indemnity structure. The second problem employs a robust optimization framework, where the policyholder aims to secure robust insurance indemnity by minimizing the worst-case convex distortion risk measure while adhering to a guaranteed VaR constraint. In this context, we provide explicit characterizations of both the optimal indemnity and the worst-case distribution in closed form through a combined approach using the Lagrangian method and modification arguments. To illustrate the practical implications of our theoretical findings, we include a concrete example based on Tail Value-at-Risk (TVaR).

q-fin.RM↗

arXiv2026

SymGPT: Auditing Smart Contracts via Combining Symbolic Execution with Large Language Models

To govern smart contracts running on Ethereum, multiple Ethereum Request for Comment (ERC) standards have been developed, each defining a set of rules governing contract behavior. Violating these rules can cause serious security issues and financial losses, signifying the importance of verifying ERC compliance. Today's practices of such verification include manual audits, expert-developed program-analysis tools, and large language models (LLMs), all of which remain ineffective at detecting ERC rule violations. This paper introduces SymGPT, a tool that combines LLMs with symbolic execution to automatically verify smart contracts' compliance with ERC rules. We begin by empirically analyzing 132 ERC rules from three major ERC standards, examining their content, security implications, and natural language descriptions. Based on this study, SymGPT instructs an LLM to translate ERC rules into a domain-specific language, synthesizes constraints from the translated rules to model potential rule violations, and performs symbolic execution for violation detection. Our evaluation shows that SymGPT identifies 5,783 ERC rule violations in 4,000 real-world contracts, including 1,375 violations with clear attack paths for financial theft. Furthermore, SymGPT outperforms six automated techniques and a security-expert auditing service, underscoring its superiority over current smart contract analysis methods.

cs.AI↗

arXiv2026

SourceBench: Can AI Answers Reference Quality Web Sources?

Large language models (LLMs) increasingly answer queries by citing web sources, but existing evaluations emphasize answer correctness rather than evidence quality. We introduce SourceBench, a benchmark for measuring the quality of cited web sources across 100 real-world queries spanning informational, factual, argumentative, social, and shopping intents. SourceBench uses an eight-metric framework covering content quality (content relevance, factual accuracy, objectivity) and page-level signals (e.g., freshness, authority/accountability, clarity), and includes a human-labeled dataset with a calibrated LLM-based evaluator that matches expert judgments closely. We evaluate eight LLMs, Google Search, and three AI search tools over 3996 cited sources using SourceBench and conduct further experiments to understand the evaluation results. Overall, our work reveals four key new insights that can guide future research in the direction of GenAI and web search.

cs.AI↗

arXiv2026

Optimal design of reinsurance contracts with a continuum of risk assessments

In this article, we employ a principal-agent model to analyze optimal contract design in a monopolistic reinsurance market under adverse selection with a continuum of insurer types. Instead of using the classical expected utility framework, we model each insurer's risk preference through their VaR at their chosen risk tolerance level. Under informational asymmetry, the reinsurer (principal) seeks to maximize expected profit by offering an optimal menu of reinsurance contracts to a continuum of insurers (agents) with hidden characteristics. To avoid the complexity of the traditional duality approach, which yields indirect multivariate utility functions, we introduce a change of variables that reduces the problem to a univariate one. We show that the optimal indirect utility for both stop-loss and quota-share reinsurance is in stop-loss form, implying that the reinsurer will classify agents into two risk groups-high and low-even in the continuum setting. Utilizing this new class of indirect utility functions, we fully solve the problem for three common reinsurance structures: stop-loss, quota-share, and change-loss. Numerical examples are also provided for illustrating the main findings.

q-fin.RM↗

arXiv2025

Beat the long tail: Distribution-Aware Speculative Decoding for RL Training

Reinforcement learning(RL) post-training has become essential for aligning large language models (LLMs), yet its efficiency is increasingly constrained by the rollout phase, where long trajectories are generated token by token. We identify a major bottleneck:the long-tail distribution of rollout lengths, where a small fraction of long generations dominates wall clock time and a complementary opportunity; the availability of historical rollouts that reveal stable prompt level patterns across training epochs. Motivated by these observations, we propose DAS, a Distribution Aware Speculative decoding framework that accelerates RL rollouts without altering model outputs. DAS integrates two key ideas: an adaptive, nonparametric drafter built from recent rollouts using an incrementally maintained suffix tree, and a length aware speculation policy that allocates more aggressive draft budgets to long trajectories that dominate makespan. This design exploits rollout history to sustain acceptance while balancing base and token level costs during decoding. Experiments on math and code reasoning tasks show that DAS reduces rollout time up to 50% while preserving identical training curves, demonstrating that distribution-aware speculative decoding can significantly accelerate RL post training without compromising learning quality.

cs.LG↗

arXiv2025

Learning Semantics, Not Addresses: Runtime Neural Prefetching for Far Memory

Memory prefetching has long boosted CPU caches and is increasingly vital for far-memory systems, where large portions of memory are offloaded to cheaper, remote tiers. While effective prefetching requires accurate prediction of future accesses, prior ML approaches have been limited to simulation or small-scale hardware. We introduce FarSight, the first Linux-based far-memory system to leverage deep learning by decoupling application semantics from runtime memory layout. This separation enables offline-trained models to predict access patterns over a compact ordinal vocabulary, which are resolved at runtime through lightweight mappings. Across four data-intensive workloads, FarSight delivers up to 3.6x higher performance than the state-of-the-art.

cs.LG↗

arXiv2025

An Early Exploration of Deep-Learning-Driven Prefetching for Far Memory

Far-memory systems, where applications store less-active data in more energy-efficient memory media, are increasingly adopted by data centers. However, applications are bottlenecked by on-demand data fetching from far- to local-memory. We present Memix, a far-memory system that embodies a deep-learning-system co-design for efficient and accurate prefetching, minimizing on-demand far-memory accesses. One key observation is that memory accesses are shaped by both application semantics and runtime context, providing an opportunity to optimize each independently. Preliminary evaluation of Memix on data-intensive workloads shows that it outperforms the state-of-the-art far-memory system by up to 42%.

cs.OS↗

arXiv2025

Field-free superconducting diode effect of NbSe2 induced by strain

Superconducting diodes, similar to semiconductor diodes, possess unidirectional superconducting properties and are the fundamental units for constructing superconducting quantum computing, thus attracting widespread attention. At present, most of superconducting diodes require an external magnetic field or proximity effect to break time reversal symmetry (TRS). The cases of intrinsic superconducting diode effect (SDE) under zero magnetic field are relatively scarce, and there are still some puzzles especially regarding the reasons for the TRS breaking. Here, we not only report field free SDE in NbSe2 induced by strain, but also large values of the difference of Ic+ and |Ic-| (ΔIc) of 286 μA and the superconducting diode efficiency (η) of 6.76 % are achieved. Interestingly, ΔIc varies with the magnetic field and exhibits two distinct evolutionary behaviors with B-odd or B-even symmetry in various devices. We attribute this to the selective activation of two independent, spatially-orthogonal mechanisms: a stress-induced real-space polarity and a field-induced reciprocal-space asymmetric energy bands. In general, we propose an extremely effectively method to produce field free SDE, even when the material itself does not possess field free SDE, and provide new perspectives to understand the SDE which build new avenues for superconducting quantum devices.

cond-mat.mtrl-sci↗

arXiv2025

How to Save My Gas Fees: Understanding and Detecting Real-world Gas Issues in Solidity Programs

The execution of smart contracts on Ethereum, a public blockchain system, incurs a fee called gas fee for its computation and data storage. When programmers develop smart contracts (e.g., in the Solidity programming language), they could unknowingly write code snippets that unnecessarily cause more gas fees. These issues, or what we call gas wastes, can lead to significant monetary losses for users. This paper takes the initiative in helping Ethereum users reduce their gas fees in two key steps. First, we conduct an empirical study on gas wastes in open-source Solidity programs and Ethereum transaction traces. Second, to validate our study findings, we develop a static tool called PeCatch to effectively detect gas wastes in Solidity programs, and manually examine the Solidity compiler's code to pinpoint implementation errors causing gas wastes. Overall, we make 11 insights and four suggestions, which can foster future tool development and programmer awareness, and fixing our detected bugs can save $0.76 million in gas fees daily.

cs.SE↗

arXiv2025

Synthesis of intrinsic magnetic topological insulator MnBi2nTe3n+1 family by chemical vapor transport method with feedback regulation

MnBi2nTe3n+1 (MBT) is a representative family of intrinsic magnetic topological insulators, in which numerous exotic phenomena such as the quantum anomalous Hall effect are expected. The high-quality crystal growth and magnetism manipulation are the most essential processes. Here we develop a modified chemical vapor transport method using a feedback-regulated strategy, which provides the closed-loop control of growth temperature within +/- 0.1 degree Celsius. Single crystals of MnBi2Te4, MnBi4Te7, and MnBi6Te10 are obtained under different temperature intervals respectively, and show variable tunability on magnetism by finely tuning the growth temperatures. Specifically, the cold-end temperatures not only vary the strength of antiferromagnetic coupling in MnBi2Te4, but also induce magnetic ground state transitions from antiferromagnetism to ferromagnetism in MnBi4Te7 and MnBi6Te10. In MnBi2Te4 with optimized magnetism, quantized transport with Chern insulator state is also realized at the low field of 3.7 T. Our results provide a systematic picture for the crystal growth and the rich magnetic tunability of MBT family, providing richer platforms for the related researches combining magnetism and topological physics.

cond-mat.mtrl-sci↗

arXiv2025

Deep Learning Model Inversion Attacks and Defenses: A Comprehensive Survey

The rapid adoption of deep learning in sensitive domains has brought tremendous benefits. However, this widespread adoption has also given rise to serious vulnerabilities, particularly model inversion (MI) attacks, posing a significant threat to the privacy and integrity of personal data. The increasing prevalence of these attacks in applications such as biometrics, healthcare, and finance has created an urgent need to understand their mechanisms, impacts, and defense methods. This survey aims to fill the gap in the literature by providing a structured and in-depth review of MI attacks and defense strategies. Our contributions include a systematic taxonomy of MI attacks, extensive research on attack techniques and defense mechanisms, and a discussion about the challenges and future research directions in this evolving field. By exploring the technical and ethical implications of MI attacks, this survey aims to offer insights into the impact of AI-powered systems on privacy, security, and trust. In conjunction with this survey, we have developed a comprehensive repository to support research on MI attacks and defenses. The repository includes state-of-the-art research papers, datasets, evaluation metrics, and other resources to meet the needs of both novice and experienced researchers interested in MI attacks and defenses, as well as the broader field of AI security and privacy. The repository will be continuously maintained to ensure its relevance and utility. It is accessible at https://github.com/overgter/Deep-Learning-Model-Inversion-Attacks-and-Defenses.

cs.CR↗

arXiv2025

On multivariate contribution measures of systemic risk with applications in cryptocurrency market

Conditional risk measures and their associated risk contribution measures are commonly employed in finance and actuarial science for evaluating systemic risk and quantifying the effects of risk interactions. This paper introduces various types of contribution ratio measures based on the MCoVaR, MCoES, and MMME studied in Ortega-Jiménez et al. (2021) and Das & Fasen-Hartmann (2018) to assess the relative effects of a single risk when other risks in a group are in distress. The properties of these contribution risk measures are examined, and sufficient conditions for comparing these measures between two sets of random vectors are established using univariate and multivariate stochastic orders and statistically dependent notions. Numerical examples are presented to validate these conditions. Finally, a real dataset from the cryptocurrency market is used to analyze the spillover effects through our proposed contribution measures.

q-fin.RM↗