arXiv subjects

Jianan Zhang

Publications and source records attributed to Jianan Zhang.

At least 19 recordsLinked to original sources

arXiv2026

Qwen-UI-Agent Technical Report: Toward Next-Generation Real-World Centric Foundation GUI Agents

GUI agents have the potential to become a general purpose executor over existing digital devices. To advance them toward real-world use, we envision agents that operate reliably on real devices, execute workflows across platforms, combine GUI interaction with CLI execution, complete long-horizon tasks, proactively initiate useful services, and autonomously improve their capabilities with minimal human effort. Guided by this vision, we present Qwen-UI-Agent, a real-world centric foundation GUI agent spanning mobile, computer-use, web, and DeepSearch environments. Qwen-UI-Agent combines diverse sandbox environments with a large-scale real-device mobile runtime. Its unified action space interleaves GUI operations with CLI execution and generates batched actions in a single model turn. An AutoResearch-style data flywheel uses agents to construct tasks and environments, diagnose failures, and plan subsequent iterations. Online RL supports training on trajectories exceeding 100 turns, with over 10,000 concurrent environments accelerating rollout. A lightweight harness layer supports proactive service initiation and stateful workflows across mobile and computer. Across a broad suite of evaluations, Qwen-UI-Agent sets state-of-the-art performance on mobile-use benchmarks while delivering competitive performance on computer- and browser-use tasks against frontier models, including Opus 4.8, Gemini 3.1 Pro, and GPT-5.6 Sol. On mobile use, it achieves 82.1% on MobileWorld, 92.2% on MobileWorld-Real, and 97.5% on AndroidDaily. On computer use, it achieves 79.5% on OSWorld-Verified and a 40.0% partial-progress score on OSWorld-v2. On browser use and GUI grounding, it achieves 73.6% on WebArena and 81.5% on ScreenSpot-Pro, respectively.

cs.AI↗

arXiv2026

Dual-stream attention-guided learning for weakly supervised whole slide image classification

Whole slide images (WSIs) play a crucial role in cancer diagnosis due to their ultra-high resolution and rich morphological information, and multiple instance learning (MIL) has become a prevalent paradigm to solve the massive size of WSIs and the scarcity of fine-grained annotations of instance. However, most existing MIL methods struggle to accurately identify diagnostically critical local regions (instance) using only slide-level labels, and suffer from modelling the relationship of instances efficiently. To address these defects, we propose a Dual-Stream Attention-Guided Learning (DSAGL) framework. DSAGL bridges slide-level supervision and instance-level learning through a teacher-student dual-stream architecture, and mitigates instance ambiguity by generating attention-guided pseudo labels. The framework employs a shared lightweight encoder to efficiently model long-range dependencies and an attention-based fusion mechanism to enhance sensitivity to sparse, informative regions. Extensive experiments on synthetic benchmarks and real-world pathological WSI datasets demonstrate that DSAGL consistently outperforms state-of-the-art MIL methods, achieving superior discriminative performance and robustness under weak supervision.

cs.CV↗

arXiv2026

Brain-Inspired Capture: Evidence-Driven Neuromimetic Perceptual Simulation for Visual Decoding

Visual decoding of neurophysiological signals is a critical challenge for brain-computer interfaces (BCIs) and computational neuroscience. However, current approaches are often constrained by the systematic and stochastic gaps between neural and visual modalities, largely neglecting the intrinsic computational mechanisms of the Human Visual System (HVS). To address this, we propose Brain-Inspired Capture (BI-Cap), a neuromimetic perceptual simulation paradigm that aligns these modalities by emulating HVS processing. Specifically, we construct a neuromimetic pipeline comprising four biologically plausible dynamic and static transformations, coupled with Mutual Information (MI)-guided dynamic blur regulation to simulate adaptive visual processing. Furthermore, to mitigate the inherent non-stationarity of neural activity, we introduce an evidence-driven latent space representation. This formulation explicitly models uncertainty, thereby ensuring robust neural embeddings. Extensive evaluations on zero-shot brain-to-image retrieval across two public benchmarks demonstrate that BI-Cap substantially outperforms state-of-the-art methods, achieving relative gains of 9.2\% and 8.0\%, respectively. We have released the source code on GitHub through the link https://github.com/flysnow1024/BI-Cap.

cs.CV↗

arXiv2025

MobileWorld: Benchmarking Autonomous Mobile Agents in Agent-User Interactive and MCP-Augmented Environments

Among existing online mobile-use benchmarks, AndroidWorld has emerged as the dominant benchmark due to its reproducible environment and deterministic evaluation; however, recent agents achieving over 90% success rates indicate its saturation and motivate the need for a more challenging benchmark. In addition, its environment lacks key application categories, such as e-commerce and enterprise communication, and does not reflect realistic mobile-use scenarios characterized by vague user instructions and hybrid tool usage. We introduce MobileWorld, a substantially more challenging benchmark designed to reflect real-world usage through 201 tasks across 20 applications. MobileWorld derives its difficulty from an emphasis on long-horizon, cross-application workflows, requiring nearly twice as many completion steps on average (27.8 vs. 14.3) and featuring a significantly higher proportion of multi-app tasks (62.2% vs. 9.5%) than AndroidWorld. To overcome the limitations of existing environments, MobileWorld achieves a balance between production-grade utility and reproducible evaluation by utilizing open-source alternatives to industry standards (e.g., Mattermost for Slack). This approach enables a fully observable and controlled environment through source code modification and direct backend database access for precise verification. MobileWorld also introduces novel task categories, including agent-user interaction and Model Context Protocol (MCP)-augmented tasks, for evaluating agents in user-aware, hybrid-tool scenarios. To facilitate evaluation, we develop a planner-executor agentic framework with extended action spaces to support user interactions and MCP calls. Our results reveal a sharp performance drop compared to AndroidWorld, with the best agentic framework and end-to-end model achieving 51.7% and 20.9% success rates, respectively, highlighting ample headroom for future research.

cs.CL↗

arXiv2025

MAI-UI Technical Report: Real-World Centric Foundation GUI Agents

The development of GUI agents could revolutionize the next generation of human-computer interaction. Motivated by this vision, we present MAI-UI, a family of foundation GUI agents spanning the full spectrum of sizes, including 2B, 8B, 32B, and 235B-A22B variants. We identify four key challenges to realistic deployment: the lack of native agent-user interaction, the limits of UI-only operation, the absence of a practical deployment architecture, and brittleness in dynamic environments. MAI-UI addresses these issues with a unified methodology: a self-evolving data pipeline that expands the navigation data to include user interaction and MCP tool calls, a native device-cloud collaboration system routes execution by task state, and an online RL framework with advanced optimizations to scale parallel environments and context length. MAI-UI establishes new state-of-the-art across GUI grounding and mobile navigation. On grounding benchmarks, it reaches 73.5% on ScreenSpot-Pro, 91.3% on MMBench GUI L2, 70.9% on OSWorld-G, and 49.2% on UI-Vision, surpassing Gemini-3-Pro and Seed1.8 on ScreenSpot-Pro. On mobile GUI navigation, it sets a new SOTA of 76.7% on AndroidWorld, surpassing UI-Tars-2, Gemini-2.5-Pro and Seed1.8. On MobileWorld, MAI-UI obtains 41.7% success rate, significantly outperforming end-to-end GUI models and competitive with Gemini-3-Pro based agentic frameworks. Our online RL experiments show significant gains from scaling parallel environments from 32 to 512 (+5.2 points) and increasing environment step budget from 15 to 50 (+4.3 points). Finally, the native device-cloud collaboration system improves on-device performance by 33%, reduces cloud model calls by over 40%, and preserves user privacy.

cs.CV↗

arXiv2025

Embodied Intelligent Wireless (EIW): Synesthesia of Machines Empowered Wireless Communications

The evolution toward the sixth-generation (6G) and beyond mobile communication systems is marked by a fundamental shift from merely connecting devices to enabling pervasive and embodied intelligence. While recent advances in artificial intelligence (AI)-native wireless communication designs have achieved remarkable progress, the prevailing paradigm remains limited to static, modular AI substitutions. This approach fails to meet the core requirements of future wireless networks: the ability to continuously perceive, adapt to, and interact with the dynamic wireless environment. To bridge this gap, this paper introduces Embodied Intelligent Wireless (EIW), a novel communication paradigm inspired by embodied intelligence, which redefines the communication node as an active, environment-aware, and evolving entity. EIW is built around an observation-decision-action paradigm, comprising: multi-dimensional observations for comprehensive awareness of the environment and system states, a unified decision module for orchestrating multiple wireless agents in an interpretable manner, and actions where wireless agents exert practical effects on both the environment and communication systems. Furthermore, two enabling technologies, wireless world models as well as self-update and self-evolution mechanisms, are introduced to support training efficiency improvement, counterfactual evaluation, and better adaptation. Unlike existing communication systems, EIW envisions future communication systems not as passive data pipelines, but as intelligent entities that continuously interact with and co-evolve alongside their environments. Finally, through simulations, we showcase the advantages of the proposed EIW paradigm and its enabling techniques in shaping the design of wireless communication nodes.

eess.SP↗

arXiv2025

Metasurface-Enabled Superheterodyne Transmitter With Decoupled Harmonic-Free Signal Generation and Precoding

The evolution of programmable metasurfaces (PM) from passive beamforming to active information transmission marks a paradigm shift for next-generation wireless systems. However, this transition is hindered by fundamental limitations in conventional metasurface transmitter architectures, including restricted modulation orders, symbol-level spatial inconsistency, and significant harmonic interference. These issues stem from the intrinsic coupling between baseband signal processing and radio-frequency beamforming in monolithic designs reliant on simplistic switching mechanisms. This paper proposes a novel metasurface-enabled superheterodyne architecture (MSA) that fundamentally decouples these functionalities. The MSA introduces a dual-stage up-conversion process, comprising a digital up-conversion module for in-phase/quadrature modulation and baseband-to-intermediate frequency conversion, a precoder module for precoding, and a custom-designed magnitude-phase-decoupled metasurface that acts as a reconfigurable reflective mixer array. This decoupling of harmonic-free waveform generation from spatial precoding overcomes the critical drawbacks of existing approaches. Experimental results from a 5.8 GHz proof-of-concept prototype system validate the MSA's superior performance. The system generates spatially isotropic constellations for arbitrary-order QAM modulations, ensures consistent time-frequency signatures for applications like Doppler-spoofing, and achieves data rates up to 20 Mbps within a linear operating region that minimizes nonlinear distortion. The capability of employing spatial diversity and multi-stream interference cancellation has been demonstrated for the first time in a PM-based transmitter.

cs.IT↗

arXiv2025

Understanding Network Behaviors through Natural Language Question-Answering

Modern large-scale networks introduce significant complexity in understanding network behaviors, increasing the risk of misconfiguration. Prior work proposed to understand network behaviors by mining network configurations, typically relying on domain-specific languages interfaced with formal models. While effective, they suffer from a steep learning curve and limited flexibility. In contrast, natural language (NL) offers a more accessible and interpretable interface, motivating recent research on NL-guided network behavior understanding. Recent advances in large language models (LLMs) further enhance this direction, leveraging their extensive prior knowledge of network concepts and strong reasoning capabilities. However, three key challenges remain: 1) numerous router devices with lengthy configuration files challenge LLM's long-context understanding ability; 2) heterogeneity across devices and protocols impedes scalability; and 3) complex network topologies and protocols demand advanced reasoning abilities beyond the current capabilities of LLMs. To tackle the above challenges, we propose NetMind, a novel framework for querying networks using NL. Our approach introduces a tree-based configuration chunking strategy to preserve semantic coherence while enabling efficient partitioning. We then construct a unified fact graph as an intermediate representation to normalize vendor-specific configurations. Finally, we design a hybrid imperative-declarative language to reduce the reasoning burden on LLMs and enhance precision. We contribute a benchmark consisting of NL question-answer pairs paired with network configurations. Experiments demonstrate that NetMind achieves accurate and scalable network behavior understanding, outperforming existing baselines.

cs.CL↗

arXiv2025

UI-Ins: Enhancing GUI Grounding with Multi-Perspective Instruction-as-Reasoning

GUI grounding, which maps natural-language instructions to actionable UI elements, is a core capability of GUI agents. Prior works largely treats instructions as a static proxy for user intent, overlooking the impact of instruction diversity and quality on grounding performance. Through a careful investigation of existing grounding datasets, we find a 23.3% flaw rate in their instructions and show that inference-time exploitation of instruction diversity yields up to a substantial 76% relative performance improvement. In this paper, we introduce the Instruction-as-Reasoning paradigm, treating instructions as dynamic analytical pathways that offer distinct perspectives and enabling the model to select the most effective pathway during reasoning. To achieve this, we propose a two-stage training framework: supervised fine-tuning (SFT) on synthesized, diverse instructions to instill multi-perspective reasoning, followed by reinforcement learning (RL) to optimize pathway selection and composition. Our resulting models, UI-Ins-7B and UI-Ins-32B, achieve state-of-the-art results on five challenging grounding benchmarks and exhibit emergent reasoning, selectively composing and synthesizing novel instruction pathways at inference. In particular, UI-Ins-32B attains the best grounding accuracy, scoring 87.3% on UI-I2E-Bench, 57.0% on ScreenSpot-Pro, and 84.9% on MMBench-GUI L2. Furthermore, our model demonstrates strong agentic potential, achieving a 74.1% success rate on AndroidWorld using UI-Ins-7B as the executor. Our in-depth analysis reveals additional insights such as how reasoning can be formulated to enhance rather than hinder grounding performance, and how our method mitigates policy collapse in the SFT+RL framework. All code and model checkpoints will be publicly released in https://github.com/alibaba/UI-Ins.

cs.CV↗

arXiv2025

Metasurface-Enabled Superheterodyne Transmitter for Arbitrary-Order Modulation with Spatially Isotropic Symbol Distribution

Electromagnetically programmable information metasurfaces, as dynamically controllable 2D metamaterials, hold significant promise as low-profile hardware enabling passive wave control and signal generation for backscatter systems. However, current metasurface-based transmitters architecture fundamentally suffer from hardware non-modularization, forcing all transmitter functions onto nonlinear switch-based unit cells, which introduces symbol mapping inconsistency via phase coupling. Moreover, both temporal coding (limited by unit cell diodes) and space-time coding (impaired by symbol anisotropy) exhibit irreducible harmonic interference and entangled control of amplitude, phase, and beam direction. This paper proposes a metasurface-enabled superheterodyne architecture (MSA), comprising a digital up-conversion (DUC) module performing baseband-to-intermediate frequency (IF) conversion, filtering, and digital-to-analog conversion (DAC), and a reconfigurable metasurface featuring programmable unit cells that independently control both the magnitude and phase of the reflection coefficient. Systematically, the architecture leverages a dual-stage up-conversion process, typical of superheterodyne systems, but uniquely employs the metasurface for the final RF conversion stage. Building upon this framework, a proof-of-concept prototype featuring a 5.8 GHz magnitude-phase decoupled (MPD) metasurface (<15 degree phase deviation per state) and a DAC-based DUC module is presented. Extensive validation confirms the metasurface's capability for distortion-free mixing with arbitrary IF signals while maintaining consistent radiation patterns. The prototype successfully implements diverse QAM modulation schemes (4QAM to 256QAM) in mono-static and bi-static configurations, demonstrating symbol isotropy for spatially separated receivers and achieving a data rate of approximately 20 Mbps (at 5 MHz IF)...

eess.SY↗

arXiv2025

FMaMIL: Frequency-Driven Mamba Multi-Instance Learning for Weakly Supervised Lesion Segmentation in Medical Images

Accurate lesion segmentation in histopathology images is essential for diagnostic interpretation and quantitative analysis, yet it remains challenging due to the limited availability of costly pixel-level annotations. To address this, we propose FMaMIL, a novel two-stage framework for weakly supervised lesion segmentation based solely on image-level labels. In the first stage, a lightweight Mamba-based encoder is introduced to capture long-range dependencies across image patches under the MIL paradigm. To enhance spatial sensitivity and structural awareness, we design a learnable frequency-domain encoding module that supplements spatial-domain features with spectrum-based information. CAMs generated in this stage are used to guide segmentation training. In the second stage, we refine the initial pseudo labels via a CAM-guided soft-label supervision and a self-correction mechanism, enabling robust training even under label noise. Extensive experiments on both public and private histopathology datasets demonstrate that FMaMIL outperforms state-of-the-art weakly supervised methods without relying on pixel-level annotations, validating its effectiveness and potential for digital pathology applications.

cs.CV↗

arXiv2025

SimDC: A High-Fidelity Device Simulation Platform for Device-Cloud Collaborative Computing

The advent of edge intelligence and escalating concerns for data privacy protection have sparked a surge of interest in device-cloud collaborative computing. Large-scale device deployments to validate prototype solutions are often prohibitively expensive and practically challenging, resulting in a pronounced demand for simulation tools that can emulate realworld scenarios. However, existing simulators predominantly rely solely on high-performance servers to emulate edge computing devices, overlooking (1) the discrepancies between virtual computing units and actual heterogeneous computing devices and (2) the simulation of device behaviors in real-world environments. In this paper, we propose a high-fidelity device simulation platform, called SimDC, which uses a hybrid heterogeneous resource and integrates high-performance servers and physical mobile phones. Utilizing this platform, developers can simulate numerous devices for functional testing cost-effectively and capture precise operational responses from varied real devices. To simulate real behaviors of heterogeneous devices, we offer a configurable device behavior traffic controller that dispatches results on devices to the cloud using a user-defined operation strategy. Comprehensive experiments on the public dataset show the effectiveness of our simulation platform and its great potential for application. The code is available at https://github.com/opas-lab/olearning-sim.

cs.DC↗

arXiv2025

Arbitrary Waveform Generated Metasurface: A New Paradigm for Direct Modulation and Beamforming Decoupling

Information Metasurface, also known as reconfigurable intelligent surface (RIS) has gained significant attention owing to its impressive abilities in electromagnetic (EM) wave manipulation with simple structures. Numerous studies focus on achieving efficient and versatile information transmission using RIS across various fields like wireless communication, radar detection, integrated sensing, and communications, among others. Previous studies demonstrate diverse approaches to achieve reflection modulation by utilizing the superposition of the quantified reflection coefficient (RC) of each unit but suffer from the computing complexity of codebook sequence, the safety of communication, and the flexibility of modulation. To address these challenges, we introduce a novel concept of information metasurface, namely AWG-RIS, which is capable of independently producing arbitrary baseband waveforms and beam patterns through a design that decouples magnitude and phase, without changing the beam pattern. The AWG-RIS functions as a reflection mixer, directly embedding the intended signal into the incoming EM waves. Subsequently, we developed an analysis framework and introduced the waveform factor and beamforming factor into the new model, offering theoretical support for the transition from the control signal to the outgoing electromagnetic wave. Additionally, we unveil the world's first prototype showcasing passive arbitrary waveform generation while maintaining the beam pattern unaltered. Leveraging the decoupling of direct modulation and beamforming, we explore additional applications in several domains relative to traditional RISs. Finally, we present experiments that confirm the generation of arbitrary waveforms and particular spectrograms.

eess.SY↗

arXiv2025

Semi-Gradient SARSA Routing with Theoretical Guarantee on Traffic Stability and Weight Convergence

We consider the traffic control problem of dynamic routing over parallel servers, which arises in a variety of engineering systems such as transportation and data transmission. We propose a semi-gradient, on-policy algorithm that learns an approximate optimal routing policy. The algorithm uses generic basis functions with flexible weights to approximate the value function across the unbounded state space. Consequently, the training process lacks Lipschitz continuity of the gradient, boundedness of the temporal-difference error, and a prior guarantee on ergodicity, which are the standard prerequisites in existing literature on reinforcement learning theory. To address this, we combine a Lyapunov approach and an ordinary differential equation-based method to jointly characterize the behavior of traffic state and approximation weights. Our theoretical analysis proves that the training scheme guarantees traffic state stability and ensures almost surely convergence of the weights to the approximate optimum. We also demonstrate via simulations that our algorithm attains significantly faster convergence than neural network-based methods with an insignificant approximation error.

cs.LG↗

arXiv2024

TRIS-HAR: Transmissive Reconfigurable Intelligent Surfaces-assisted Cognitive Wireless Human Activity Recognition Using State Space Models

Human activity recognition (HAR) using radio frequency (RF) signals has garnered considerable attention for its applications in smart environments. However, traditional systems often struggle with limited independent channels between transmitters and receivers, multipath fading, and environmental noise, which particularly degrades performance in through-the-wall scenarios. In this paper, we present a transmissive reconfigurable intelligent surface (TRIS)-assisted through-the-wall human activity recognition (TRIS-HAR) system. The system employs TRIS technology to actively reshape wireless signal propagation, creating multiple independent paths to enhance signal clarity and improve recognition accuracy in complex indoor settings. Additionally, we propose the Human intelligence Mamba (HiMamba), an advanced state space model that captures temporal and frequency-based information for precise activity recognition. HiMamba achieves state-of-the-art performance on two public datasets, demonstrating superior accuracy. Extensive experiments indicate that the TRIS-HAR system improves recognition performance from 85.00% to 98.06% in laboratory conditions and maintains high performance across various environments. This approach offers a robust solution for enhancing RF-based HAR, with promising applications in smart home and elderly care systems.

eess.SP↗

arXiv2024

Learning-Based Adaptive Dynamic Routing with Stability Guarantee for a Single-Origin-Single-Destination Network

We consider learning-based adaptive dynamic routing for a single-origin-single-destination queuing network with stability guarantees. Specifically, we study a class of generalized shortest path policies that can be parameterized by only two constants via a piecewise-linear function. Using the Foster-Lyapunov stability theory, we develop a criterion on the parameters to ensure mean boundedness of the traffic state. Then, we develop a policy iteration algorithm that learns the parameters from realized sample paths. Importantly, the piecewise-linear function is both integrated into the Lyapunov function for stability analysis and used as a proxy of the value function for policy iteration; hence, stability is inherently ensured for the learned policy. Finally, we demonstrate via a numerical example that the proposed algorithm learns a near-optimal routing policy with an acceptable optimality gap but significantly higher computational efficiency compared with a standard neural network-based algorithm.

eess.SY↗

arXiv2024

Measurement of microwave polarization using two polarization orthogonal local microwave electric fields in a Rydberg atom-based mixer

We propose and demonstrate a novel method for measuring the polarization direction of a microwave electric field in a single measurement using a Rydberg atom-based mixer with two orthogonally polarized local microwave electric fields. Furthermore, introducing a weak static magnetic field enables the utilization of the Zeeman effect and exploitation of polarization asymmetry. This distinction allows for determining the polarization direction of the microwave field isθor180°-θwithin the 0 to 180 degree range. This is the first real-time measurement of microwave polarization within 0 to 180 degrees, crucial for microwave sensing and information transmission.

physics.atom-ph↗

arXiv2024

TRGR: Transmissive RIS-aided Gait Recognition Through Walls

Gait recognition with radio frequency (RF) signals enables many potential applications requiring accurate identification. However, current systems require individuals to be within a line-of-sight (LOS) environment and struggle with low signal-to-noise ratio (SNR) when signals traverse concrete and thick walls. To address these challenges, we present TRGR, a novel transmissive reconfigurable intelligent surface (RIS)-aided gait recognition system. TRGR can recognize human identities through walls using only the magnitude measurements of channel state information (CSI) from a pair of transceivers. Specifically, by leveraging transmissive RIS alongside a configuration alternating optimization algorithm, TRGR enhances wall penetration and signal quality, enabling accurate gait recognition. Furthermore, a residual convolution network (RCNN) is proposed as the backbone network to learn robust human information. Experimental results confirm the efficacy of transmissive RIS, highlighting the significant potential of transmissive RIS in enhancing RF-based gait recognition systems. Extensive experiment results show that TRGR achieves an average accuracy of 97.88\% in identifying persons when signals traverse concrete walls, demonstrating the effectiveness and robustness of TRGR.

cs.AI↗