arXiv subjects

Xiao Zhang

Publications and source records attributed to Xiao Zhang.

At least 235 records · Page 13Linked to original sources

arXiv2025

PMU-Data: Data Traces Could be Distinguished

Modern processors widely equip the Performance Monitoring Unit (PMU) to collect various architecture and microarchitecture events. Software developers often utilize the PMU to enhance program's performance, but the potential side effects that arise from its activation are often disregarded. In this paper, we find that the PMU can be employed to retrieve instruction operands. Based on this discovery, we introduce PMU-Data, a novel category of side-channel attacks aimed at leaking secret by identifying instruction operands with PMU. To achieve the PMU-Data attack, we develop five gadgets to encode the confidential data into distinct data-related traces while maintaining the control-flow unchanged. We then measure all documented PMU events on three physical machines with different processors while those gadgets are performing. We successfully identify two types of vulnerable gadgets caused by DIV and MOV instructions. Additionally, we discover 40 vulnerable PMU events that can be used to carry out the PMU-Data attack. We through real experiments to demonstrate the perniciousness of the PMU-Data attack by implementing three attack goals: (1) leaking the kernel data illegally combined with the transient execution vulnerabilities including Meltdown, Spectre, and Zombieload; (2) building a covert-channel to secretly transfer data; (3) extracting the secret data protected by the Trusted Execution Environment (TEE) combined with the Zombieload vulnerability.

cs.CR↗

arXiv2025

UCFE: A User-Centric Financial Expertise Benchmark for Large Language Models

This paper introduces the UCFE: User-Centric Financial Expertise benchmark, an innovative framework designed to evaluate the ability of large language models (LLMs) to handle complex real-world financial tasks. UCFE benchmark adopts a hybrid approach that combines human expert evaluations with dynamic, task-specific interactions to simulate the complexities of evolving financial scenarios. Firstly, we conducted a user study involving 804 participants, collecting their feedback on financial tasks. Secondly, based on this feedback, we created our dataset that encompasses a wide range of user intents and interactions. This dataset serves as the foundation for benchmarking 11 LLMs services using the LLM-as-Judge methodology. Our results show a significant alignment between benchmark scores and human preferences, with a Pearson correlation coefficient of 0.78, confirming the effectiveness of the UCFE dataset and our evaluation approach. UCFE benchmark not only reveals the potential of LLMs in the financial domain but also provides a robust framework for assessing their performance and user satisfaction.

q-fin.CP↗

arXiv2025

Unsupervised Patch-GAN with Targeted Patch Ranking for Fine-Grained Novelty Detection in Medical Imaging

Detecting novel anomalies in medical imaging is challenging due to the limited availability of labeled data for rare abnormalities, which often display high variability and subtlety. This challenge is further compounded when small abnormal regions are embedded within larger normal areas, as whole-image predictions frequently overlook these subtle deviations. To address these issues, we propose an unsupervised Patch-GAN framework designed to detect and localize anomalies by capturing both local detail and global structure. Our framework first reconstructs masked images to learn fine-grained, normal-specific features, allowing for enhanced sensitivity to minor deviations from normality. By dividing these reconstructed images into patches and assessing the authenticity of each patch, our approach identifies anomalies at a more granular level, overcoming the limitations of whole-image evaluation. Additionally, a patch-ranking mechanism prioritizes regions with higher abnormal scores, reinforcing the alignment between local patch discrepancies and the global image context. Experimental results on the ISIC 2016 skin lesion and BraTS 2019 brain tumor datasets validate our framework's effectiveness, achieving AUCs of 95.79% and 96.05%, respectively, and outperforming three state-of-the-art baselines.

cs.CV↗

arXiv2025

An Item is Worth a Prompt: Versatile Image Editing with Disentangled Control

Building on the success of text-to-image diffusion models (DPMs), image editing is an important application to enable human interaction with AI-generated content. Among various editing methods, editing within the prompt space gains more attention due to its capacity and simplicity of controlling semantics. However, since diffusion models are commonly pretrained on descriptive text captions, direct editing of words in text prompts usually leads to completely different generated images, violating the requirements for image editing. On the other hand, existing editing methods usually consider introducing spatial masks to preserve the identity of unedited regions, which are usually ignored by DPMs and therefore lead to inharmonic editing results. Targeting these two challenges, in this work, we propose to disentangle the comprehensive image-prompt interaction into several item-prompt interactions, with each item linked to a special learned prompt. The resulting framework, named D-Edit, is based on pretrained diffusion models with cross-attention layers disentangled and adopts a two-step optimization to build item-prompt associations. Versatile image editing can then be applied to specific items by manipulating the corresponding prompts. We demonstrate state-of-the-art results in four types of editing operations including image-based, text-based, mask-based editing, and item removal, covering most types of editing applications, all within a single unified framework. Notably, D-Edit is the first framework that can (1) achieve item editing through mask editing and (2) combine image and text-based editing. We demonstrate the quality and versatility of the editing results for a diverse collection of images through both qualitative and quantitative evaluations.

cs.CV↗

arXiv2025

ReDeEP: Detecting Hallucination in Retrieval-Augmented Generation via Mechanistic Interpretability

Retrieval-Augmented Generation (RAG) models are designed to incorporate external knowledge, reducing hallucinations caused by insufficient parametric (internal) knowledge. However, even with accurate and relevant retrieved content, RAG models can still produce hallucinations by generating outputs that conflict with the retrieved information. Detecting such hallucinations requires disentangling how Large Language Models (LLMs) utilize external and parametric knowledge. Current detection methods often focus on one of these mechanisms or without decoupling their intertwined effects, making accurate detection difficult. In this paper, we investigate the internal mechanisms behind hallucinations in RAG scenarios. We discover hallucinations occur when the Knowledge FFNs in LLMs overemphasize parametric knowledge in the residual stream, while Copying Heads fail to effectively retain or integrate external knowledge from retrieved content. Based on these findings, we propose ReDeEP, a novel method that detects hallucinations by decoupling LLM's utilization of external context and parametric knowledge. Our experiments show that ReDeEP significantly improves RAG hallucination detection accuracy. Additionally, we introduce AARF, which mitigates hallucinations by modulating the contributions of Knowledge FFNs and Copying Heads.

cs.CL↗

arXiv2025

Modeling the Saddle-like GeV-TeV Spectrum of HESS J1809-193: $γ$-Rays Arising from Reverse-Shocked Pulsar Wind Nebula?

Evolution of pulsar wind nebulae (PWNe) could be expected to leave imprints in $γ$-rays. We suggest that intriguing GeV-TeV spectral energy distribution (SED) of HESS J1809$-$193 and Fermi-LAT source J1810.3$-$1925e is very likely to be the $γ$-ray signature of PWN J1809$-$193 in light of the scenario that the PWN was struck by the reverse shock of the parent supernova remnant. Based on evolutionary theory of PWNe, we consider that, when the PWN was disrupted during collision by the reverse shock, some very high-energy electrons escaped impulsively. The remaining electrons stayed in the relic PWN, which was displaced from the pulsar. The very high-energy part of the remaining electrons were depleted by the strong magnetic field that was enhanced by the reverse shock compression in the reverberation stage, leaving the other part of them generating GeV emission. The particles injected from the pulsar after the disruption enter the relic PWN through the newly formed tunnel called the cocoon. The $γ$-ray emission from the escaped electrons can account for the TeV spectrum of component A of HESS J1809$-$193 or the TeV halo, while the electrons remaining after disruption can account for the GeV spectrum of J1810.3$-$1925e. Thus, combination of contributions from these two populations of electrons naturally reproduces the saddle-like SED of HESS J1809$-$193 and J1810.3$-$1925e from 5 GeV to 30 TeV, together with the spectral hardening around 100 GeV. We also show that the post-disruption injection of electrons can explain the spectrum of the relatively faint $γ$-ray emission of component B of HESS J1809$-$193.

astro-ph.HE↗

arXiv2025

Generating Less Certain Adversarial Examples Improves Robust Generalization

This paper revisits the robust overfitting phenomenon of adversarial training. Observing that models with better robust generalization performance are less certain in predicting adversarially generated training inputs, we argue that overconfidence in predicting adversarial examples is a potential cause. Therefore, we hypothesize that generating less certain adversarial examples improves robust generalization, and propose a formal definition of adversarial certainty that captures the variance of the model's predicted logits on adversarial examples. Our theoretical analysis of synthetic distributions characterizes the connection between adversarial certainty and robust generalization. Accordingly, built upon the notion of adversarial certainty, we develop a general method to search for models that can generate training-time adversarial inputs with reduced certainty, while maintaining the model's capability in distinguishing adversarial examples. Extensive experiments on image benchmarks demonstrate that our method effectively learns models with consistently improved robustness and mitigates robust overfitting, confirming the importance of generating less certain adversarial examples for robust generalization. Our implementations are available as open-source code at: https://github.com/TrustMLRG/AdvCertainty.

cs.LG↗

arXiv2025

ReARTeR: Retrieval-Augmented Reasoning with Trustworthy Process Rewarding

Retrieval-Augmented Generation (RAG) systems for Large Language Models (LLMs) hold promise in knowledge-intensive tasks but face limitations in complex multi-step reasoning. While recent methods have integrated RAG with chain-of-thought reasoning or test-time search using Process Reward Models (PRMs), these approaches encounter challenges such as a lack of explanations, bias in PRM training data, early-step bias in PRM scores, and insufficient post-training optimization of reasoning potential. To address these issues, we propose Retrieval-Augmented Reasoning through Trustworthy Process Rewarding (ReARTeR), a framework that enhances RAG systems' reasoning capabilities through post-training and test-time scaling. At test time, ReARTeR introduces Trustworthy Process Rewarding via a Process Reward Model for accurate scalar scoring and a Process Explanation Model (PEM) for generating natural language explanations, enabling step refinement. During post-training, it utilizes Monte Carlo Tree Search guided by Trustworthy Process Rewarding to collect high-quality step-level preference data, optimized through Iterative Preference Optimization. ReARTeR addresses three core challenges: (1) misalignment between PRM and PEM, tackled through off-policy preference learning; (2) bias in PRM training data, mitigated by balanced annotation methods and stronger annotations for challenging examples; and (3) early-step bias in PRM, resolved through a temporal-difference-based look-ahead search strategy. Experimental results on multi-step reasoning benchmarks demonstrate significant improvements, underscoring ReARTeR's potential to advance the reasoning capabilities of RAG systems.

cs.CL↗

arXiv2025

Expanding Molecular Shell and Possible γ-ray Source Associated with Supernova Remnant Kesteven 67

We investigate the molecular environment of the supernova remnant (SNR) Kesteven 67 (G18.8+0.3) using observations in $^{12}$CO, $^{13}$CO, HCO$^+$,and HCN lines and possible associated $γ$-ray emission using 16-yr Fermi-LAT observation. We find that the SNR is closely surrounded by a molecular belt in the southeastern boundary, with the both recessed in the band-like molecular gas structure along the Galactic plane. The asymmetric molecular line profiles are widely present in the surrounding gas around local-standard-of-rest velocity +20 km s$^{-1}$. The secondary components centered at $\sim$+16km s$^{-1}$ in the belt and $\sim$+26 km s$^{-1}$ in the northern clump can be ascribed to the motion of a wind-blown molecular shell. This explanation is supported by the position-velocity diagram along a line cutting across the remnant, which shows an arc-like pattern, suggesting an expanding gas structure. With the simulation of chemical effects of shock propagation, the abundance ratios $N$(HCO$^+$)/$N$($^{12}$CO) $\sim2.6\times 10^{-5}$--$3.6\times 10^{-4}$ obtained in the belt can be more naturally interpreted by the wind-driven bubble shock than by the SNR shock. The belt and northern clump are very likely to be parts of an incomplete molecular shell of bubble driven by O-type progenitor star's wind. The analysis of 0.2--500 GeV $γ$-ray emission uncovers a possible point source (`Source~A') about 6.5$σ$ located in the north of the SNR, which essentially corresponds to northern molecular clump. Our spectral fit of the emission indicates that a hadronic origin is favored by the measured Galactic number ratio between CR electrons and protons $\sim0.01$.

astro-ph.HE↗

arXiv2025

Building a human-like observer using deep learning in an extended Wigner's friend experiment

There has been a longstanding demand for artificial intelligence with human-level cognitive sophistication to address loopholes in Bell-type experiments. In this study, we propose a novel experimental framework that integrates advanced deep learning techniques, employing neural network-based artificial intelligence in an extended Wigner's friend experiment. We demonstrate the framework through simulations and introduce three new analytical metrics-morphing polygons, averaged Shannon entropy, and probability density maps-to evaluate the results. These results can be used to determine whether our artificial intelligence qualifies as a bona fide observer and whether superposition applies to macroscopic systems, including observers.

physics.gen-ph↗

arXiv2024

Optical Routing via High Efficiency Composite Acoustic Diffraction

Acousto-optical modulation (AOM) is a powerful and widely used technique for rapidly controlling the frequency, phase, intensity, and direction of light. Based on Bragg diffraction, AOMs typically exhibit moderate diffraction efficiency, often less than 90\% even for collimated inputs. In this work, we demonstrate that this efficiency can be significantly improved using a composite (CP) setup comprising a pair of 4-F-linked AOMs, enabling beamsplitting with fully tunable splitting amplitude and phase. The efficiency enhancement arises from two effects, termed "momentum echo" and "high-order rephasing," which can be simultaneously optimized by adjusting the relative distance between the two AOMs. This method is resource-efficient, does not require ultra-collimation, and maintains control bandwidth. Experimentally, we achieved a diffraction efficiency exceeding 99\% (excluding insertion loss) and a 35 dB single-mode suppression of the 0th-order beam, demonstrating a full-contrast optical router with a switching time of less than 100~nanoseconds. Theoretically, we formulate the dynamics of CP-AOM in terms of multi-mode quantum control and discuss extensions beyond the $N=2$ configuration presented in this work. The substantially enhanced performance of CP-AOMs, coupled with reduced acoustic amplitude requirements, may significantly advance our ability to accurately control light at high speeds with low-loss acousto-optics.

physics.optics↗

arXiv2024

HELPNet: Hierarchical Perturbations Consistency and Entropy-guided Ensemble for Scribble Supervised Medical Image Segmentation

Creating fully annotated labels for medical image segmentation is prohibitively time-intensive and costly, emphasizing the necessity for innovative approaches that minimize reliance on detailed annotations. Scribble annotations offer a more cost-effective alternative, significantly reducing the expenses associated with full annotations. However, scribble annotations offer limited and imprecise information, failing to capture the detailed structural and boundary characteristics necessary for accurate organ delineation. To address these challenges, we propose HELPNet, a novel scribble-based weakly supervised segmentation framework, designed to bridge the gap between annotation efficiency and segmentation performance. HELPNet integrates three modules. The Hierarchical perturbations consistency (HPC) module enhances feature learning by employing density-controlled jigsaw perturbations across global, local, and focal views, enabling robust modeling of multi-scale structural representations. Building on this, the Entropy-guided pseudo-label (EGPL) module evaluates the confidence of segmentation predictions using entropy, generating high-quality pseudo-labels. Finally, the structural prior refinement (SPR) module incorporates connectivity and bounded priors to enhance the precision and reliability and pseudo-labels. Experimental results on three public datasets ACDC, MSCMRseg, and CHAOS show that HELPNet significantly outperforms state-of-the-art methods for scribble-based weakly supervised segmentation and achieves performance comparable to fully supervised methods. The code is available at https://github.com/IPMI-NWU/HELPNet.

cs.CV↗

arXiv2024

Do Parameters Reveal More than Loss for Membership Inference?

Membership inference attacks are used as a key tool for disclosure auditing. They aim to infer whether an individual record was used to train a model. While such evaluations are useful to demonstrate risk, they are computationally expensive and often make strong assumptions about potential adversaries' access to models and training environments, and thus do not provide tight bounds on leakage from potential attacks. We show how prior claims around black-box access being sufficient for optimal membership inference do not hold for stochastic gradient descent, and that optimal membership inference indeed requires white-box access. Our theoretical results lead to a new white-box inference attack, IHA (Inverse Hessian Attack), that explicitly uses model parameters by taking advantage of computing inverse-Hessian vector products. Our results show that both auditors and adversaries may be able to benefit from access to model parameters, and we advocate for further research into white-box methods for membership inference.

cs.LG↗

arXiv2024

Reconciling Human Development and Giant Panda Protection Goals: Cost-efficiency Evaluation of Farmland Reverting and Energy Substitution Programs in Wolong National Reserve

Balancing human development with conservation necessitates ecological policies that optimize outcomes within limited budgets, highlighting the importance of cost-efficiency and local impact analysis. This study employs the Socio-Econ-Ecosystem Multipurpose Simulator (SEEMS), an Agent-Based Model (ABM) designed for simulating small-scale Coupled Human and Nature Systems (CHANS), to evaluate the cost-efficiency of two major ecology conservation programs: Grain-to-Green (G2G) and Firewood-to-Electricity (F2E). Focusing on China Wolong National Reserve, a worldwide hot spot for flagship species conservation, the study evaluates the direct benefits of these programs, including reverted farmland area and firewood consumption, along with their combined indirect benefits on habitat quality, carbon emissions, and gross economic benefits. The findings are as follows: (1) The G2G program achieves optimal financial efficiency at approximately 500 CNY/Mu, with diminishing returns observed beyond 1000 CNY/Mu; (2) For the F2E program, the most fiscally cost-efficient option arises when the subsidized electricity price is at 0.4-0.5 CNY/kWh, while further reductions of the prices to below 0.1 CNY/kWh result in a diminishing cost-benefit ratio; (3) Comprehensive cost-efficiency analysis reveals no significant link between financial burden and carbon emissions, but a positive correlation with habitat quality and an inverted U-shaped relationship with total economic income; (4) Pareto analysis identifies 18 optimal dual-policy combinations for balancing carbon footprint, habitat quality, and gross economic benefits; (5) Posterior Pareto optimization further refines the selection of a specific policy scheme for a given realistic scenario. The analytical framework of this paper helps policymakers design economically viable and environmentally sustainable policies, addressing global conservation challenges.

cs.CY↗

arXiv2024

Trigger$^3$: Refining Query Correction via Adaptive Model Selector

In search scenarios, user experience can be hindered by erroneous queries due to typos, voice errors, or knowledge gaps. Therefore, query correction is crucial for search engines. Current correction models, usually small models trained on specific data, often struggle with queries beyond their training scope or those requiring contextual understanding. While the advent of Large Language Models (LLMs) offers a potential solution, they are still limited by their pre-training data and inference cost, particularly for complex queries, making them not always effective for query correction. To tackle these, we propose Trigger$^3$, a large-small model collaboration framework that integrates the traditional correction model and LLM for query correction, capable of adaptively choosing the appropriate correction method based on the query and the correction results from the traditional correction model and LLM. Trigger$^3$ first employs a correction trigger to filter out correct queries. Incorrect queries are then corrected by the traditional correction model. If this fails, an LLM trigger is activated to call the LLM for correction. Finally, for queries that no model can correct, a fallback trigger decides to return the original query. Extensive experiments demonstrate Trigger$^3$ outperforms correction baselines while maintaining efficiency.

cs.CL↗

arXiv2024

Nested Diffusion Models Using Hierarchical Latent Priors

We introduce nested diffusion models, an efficient and powerful hierarchical generative framework that substantially enhances the generation quality of diffusion models, particularly for images of complex scenes. Our approach employs a series of diffusion models to progressively generate latent variables at different semantic levels. Each model in this series is conditioned on the output of the preceding higher-level models, culminating in image generation. Hierarchical latent variables guide the generation process along predefined semantic pathways, allowing our approach to capture intricate structural details while significantly improving image quality. To construct these latent variables, we leverage a pre-trained visual encoder, which learns strong semantic visual representations, and modulate its capacity via dimensionality reduction and noise injection. Across multiple datasets, our system demonstrates significant enhancements in image quality for both unconditional and class/text conditional generation. Moreover, our unconditional generation system substantially outperforms the baseline conditional system. These advancements incur minimal computational overhead as the more abstract levels of our hierarchy work with lower-dimensional representations.

cs.CV↗

arXiv2024

Joint Mode Selection and Beamforming Designs for Hybrid-RIS Assisted ISAC Systems

This paper considers a hybrid reconfigurable intelligent surface (RIS) assisted integrated sensing and communication (ISAC) system, where each RIS element can flexibly switch between the active and passive modes. Subject to the signal-to-interference-plus-noise ratio (SINR) constraint for each communication user (CU) and the transmit power constraints for both the base station (BS) and the active RIS elements, with the objective of maximizing the minimum beampattern gain among multiple targets, we jointly optimize the BS transmit beamforming for ISAC and the mode selection of each RIS reflecting element, as well as the RIS reflection coefficient matrix. Such formulated joint hybrid-RIS assisted ISAC design problem is a mixed-integer nonlinear program, which is decomposed into two low-dimensional subproblems being solved in an alternating manner. Specifically, by using the semidefinite relaxation (SDR) technique along with the rank-one beamforming construction process, we efficiently obtain the optimal ISAC transmit beamforming design at the BS. Via the SDR and successive convex approximation (SCA) techniques, we jointly determine the active/passive mode selection and reflection coefficient for each RIS element. Numerical results demonstrate that the proposed design solution is significantly superior to the existing baseline solutions.

cs.IT↗

arXiv2024

Brownian spin-locking effect

Brownian systems are characterized by spatiotemporal disorder, which arises from the erratic motion of particles driven by thermal fluctuations. When light interacts with such systems, it typically produces unpolarized and uncorrelated fields. Here, we report the observation of a large-scale spin-locking effect of light within a Brownian medium. In an observation direction perpendicular to the incident wave momentum, scattering naturally divides into two diffusion regions, each associated with an opposite spin from the Brownian nanoparticles. This effect arises from the intrinsic spin-orbit interactions of scattering from individual nanoparticles, which ubiquitously generate radiative spin fields that propagate through the Brownian medium with multiple incoherent scattering. It offers a novel experimental platform for exploring macroscale spin behaviors of diffused light, with potential applications in precision metrology for measuring various nanoparticle properties. Our findings may inspire the study of analogous phenomena for different waves from novel spin-orbit interactions in complex disordered systems.

physics.optics↗