arXiv subjects

Jin Ma

Publications and source records attributed to Jin Ma.

At least 37 records · Page 2Linked to original sources

arXiv2026

UI-Copilot: Advancing Long-Horizon GUI Automation via Tool-Integrated Policy Optimization

MLLM-based GUI agents have demonstrated strong capabilities in complex user interface interaction tasks. However, long-horizon scenarios remain challenging, as these agents are burdened with tasks beyond their intrinsic capabilities, suffering from memory degradation, progress confusion, and math hallucination. To address these challenges, we present UI-Copilot, a collaborative framework where the GUI agent focuses on task execution while a lightweight copilot provides on-demand assistance for memory retrieval and numerical computation. We introduce memory decoupling to separate persistent observations from transient execution context, and train the policy agent to selectively invoke the copilot as Retriever or Calculator based on task demands. To enable effective tool invocation learning, we propose Tool-Integrated Policy Optimization (TIPO), which separately optimizes tool selection through single-turn prediction and task execution through on-policy multi-turn rollouts. Experimental results show that UI-Copilot-7B achieves state-of-the-art performance on challenging MemGUI-Bench, outperforming strong 7B-scale GUI agents such as GUI-Owl-7B and UI-TARS-1.5-7B. Moreover, UI-Copilot-7B delivers a 17.1% absolute improvement on AndroidWorld over the base Qwen model, highlighting UI-Copilot's strong generalization to real-world GUI tasks.

cs.LG↗

arXiv2026

HD-VGGT: High-Resolution Visual Geometry Transformer

High-resolution imagery is essential for accurate 3D reconstruction, as many geometric details only emerge at fine spatial scales. Recent feed-forward approaches, such as the Visual Geometry Grounded Transformer (VGGT), have demonstrated the ability to infer scene geometry from large collections of images in a single forward pass. However, scaling these models to high-resolution inputs remains challenging: the number of tokens in transformer architectures grows rapidly with both image resolution and the number of views, leading to prohibitive computational and memory costs. Moreover, we observe that visually ambiguous regions, such as repetitive patterns, weak textures, or specular surfaces, often produce unstable feature tokens that degrade geometric inference, especially at higher resolutions. We introduce HD-VGGT, a dual-branch architecture for efficient and robust high-resolution 3D reconstruction. A low-resolution branch predicts a coarse, globally consistent geometry, while a high-resolution branch refines details via a learned feature upsampling module. To handle unstable tokens, we propose Feature Modulation, which suppresses unreliable features early in the transformer. HD-VGGT leverages high-resolution images and supervision without full-resolution transformer costs, achieving state-of-the-art reconstruction quality.

cs.CV↗

arXiv2026

Robust 4D Visual Geometry Transformer with Uncertainty-Aware Priors

Reconstructing dynamic 4D scenes is an important yet challenging task. While 3D foundation models like VGGT excel in static settings, they often struggle with dynamic sequences where motion causes significant geometric ambiguity. To address this, we present a framework designed to disentangle dynamic and static components by modeling uncertainty across different stages of the reconstruction process. Our approach introduces three synergistic mechanisms: (1) Entropy-Guided Subspace Projection, which leverages information-theoretic weighting to adaptively aggregate multi-head attention distributions, effectively isolating dynamic motion cues from semantic noise; (2) Local-Consistency Driven Geometry Purification, which enforces spatial continuity via radius-based neighborhood constraints to eliminate structural outliers; and (3) Uncertainty-Aware Cross-View Consistency, which formulates multi-view projection refinement as a heteroscedastic maximum likelihood estimation problem, utilizing depth confidence as a probabilistic weight. Experiments on dynamic benchmarks show that our approach outperforms current state-of-the-art methods, reducing Mean Accuracy error by 13.43\% and improving segmentation F-measure by 10.49\%. Our framework maintains the efficiency of feed-forward inference and requires no task-specific fine-tuning or per-scene optimization.

cs.CV↗

arXiv2026

Understanding the Risks of Asphalt Art to the Reliability of Vision-Based Perception Systems

Artistic crosswalks featuring asphalt art, introduced by different organizations in recent years, aim to enhance the visibility and safety of pedestrians. However, their visual complexity may interfere with surveillance systems that rely on vision-based object detection models. In this study, we investigate the impact of asphalt art on pedestrian detection performance of a pretrained vision-based object detection model. We construct realistic crosswalk scenarios by compositing various street art patterns into a fixed surveillance scene and evaluate the model's performance in detecting pedestrians on asphalt-arted crosswalks under both benign and adversarial conditions. A benign case refers to pedestrian crosswalks painted with existing normal asphalt art, whereas an adversarial case involves digitally crafted or altered asphalt art perpetrated by an attacker. Our results show that while simple, color-based designs have minimal effect, complex artistic patterns, particularly those with high visual salience, can significantly degrade pedestrian detection performance. Furthermore, we demonstrate that adversarially crafted asphalt art can be exploited to deliberately obscure real pedestrians or generate non-existent pedestrian detections. These findings highlight a potential vulnerability in urban vision-based pedestrian surveillance systems, and underscore the importance of accounting for environmental visual variations when designing robust pedestrian perception models.

cs.CV↗

arXiv2026

Reinforcement Learning for optimal dividend problem under diffusion model

In this paper, we study the optimal dividend problem under the continuous time diffusion model with the bounded dividend rate from the Reinforcement Learning (RL) perspective. Unlike the standard literature, our main focus will be on numerical algorithms that allow part or all of the system parameters to be unspecified so that the optimal control cannot be explicitly determined. Following the RL literature we introduce the entropy-regularized exploratory control problem, which randomizes the control actions and balances the levels of exploitation and exploration, and carry out a theoretical analysis of the associated Policy Improvement (PI) and Policy Evaluation (PE) devices and the corresponding sequence of the approximating optimal strategies. Specifically, our algorithm will be based on two independent neural networks that approximate the value function and its derivative simultaneously. Such an algorithm, to the best of our knowledge, is new in the context of the optimal dividend problems, and can be effective even for the situation when the premium and/or interest rate is state dependent, hence beyond reach of the standard statistical methods. Some numerical experiments are presented to empirically demonstrate the effectiveness of our RL algorithm.

math.OC↗

arXiv2026

DisPatch: Disarming Adversarial Patches in Object Detection with Diffusion Models

Object detection is fundamental to various real-world applications, such as security monitoring and surveillance video analysis. Despite their advancements, state-of-the-art object detectors are still vulnerable to adversarial patch attacks, which can be easily applied to real-world objects to either conceal actual items or create non-existent ones, leading to severe consequences. In this work, we introduce DisPatch, the first diffusion-based defense framework for object detection. Unlike previous works that aim to "detect and remove" adversarial patches, DisPatch adopts a "regenerate and rectify" strategy, leveraging generative models to disarm attack effects while preserving the integrity of the input image. Specifically, we utilize the in-distribution generative power of diffusion models to regenerate the entire image, aligning it with benign data. A rectification process is then employed to identify and replace adversarial regions with their regenerated benign counterparts. DisPatch is attack-agnostic and requires no prior knowledge of the existing patches. Extensive experiments across multiple detectors demonstrate that DisPatch consistently outperforms state-of-the-art defenses on both hiding attacks and creating attacks, achieving the best overall mAP@0.5 score of 89.3% on hiding attacks, and lowering the attack success rate to 24.8% on untargeted creating attacks. Moreover, it strikes the balance between effectiveness and efficiency, and maintains strong robustness against adaptive attacks, making it a practical and reliable defense method.

cs.CV↗

arXiv2026

A Large-Scale Remote Sensing Dataset and VLM-based Algorithm for Fine-Grained Road Hierarchy Classification

In this work, we present SYSU-HiRoads, a large-scale hierarchical road dataset, and RoadReasoner, a vision-language-geometry framework for automatic multi-grade road mapping from remote sensing imagery. SYSU-HiRoads is built from GF-2 imagery covering 3631 km2 in Henan Province, China, and contains 1079 image tiles at 0.8 m spatial resolution. Each tile is annotated with dense road masks, vectorized centerlines, and three-level hierarchy labels, enabling the joint training and evaluation of segmentation, topology reconstruction, and hierarchy classification. Building on this dataset, RoadReasoner is designed to generate robust road surface masks, topology-preserving road networks, and semantically coherent hierarchy assignments. We strengthen road feature representation and network connectivity by explicitly enhancing frequency-sensitive cues and multi-scale context. Moreover, we perform hierarchy inference at the skeleton-segment level with geometric descriptors and geometry-aware textual prompts, queried by vision-language models to obtain linguistically interpretable grade decisions. Experiments on SYSU-HiRoads and the CHN6-CUG dataset show that RoadReasoner surpasses state-of-the-art road extraction baselines and produces accurate and semantically consistent road hierarchy maps with 72.6% OA, 64.2% F1 score, and 60.6% SegAcc. The dataset and code will be publicly released to support automated transport infrastructure mapping, road inventory updating, and broader infrastructure management applications.

cs.CV↗

arXiv2026

Omni IIE Bench: Benchmarking the Practical Capabilities of Image Editing Models

While Instruction-based Image Editing (IIE) has achieved significant progress, existing benchmarks pursue task breadth via mixed evaluations. This paradigm obscures a critical failure mode crucial in professional applications: the inconsistent performance of models across tasks of varying semantic scales. To address this gap, we introduce Omni IIE Bench, a high-quality, human-annotated benchmark specifically designed to diagnose the editing consistency of IIE models in practical application scenarios. Omni IIE Bench features an innovative dual-track diagnostic design: (1) Single-turn Consistency, comprising shared-context task pairs of attribute modification and entity replacement; and (2) Multi-turn Coordination, involving continuous dialogue tasks that traverse semantic scales. The benchmark is constructed via an exceptionally rigorous multi-stage human filtering process, incorporating a quality standard enforced by computer vision graduate students and an industry relevance review conducted by professional designers. We perform a comprehensive evaluation of 8 mainstream IIE models using Omni IIE Bench. Our analysis quantifies, for the first time, a prevalent performance gap: nearly all models exhibit a significant performance degradation when transitioning from low-semantic-scale to high-semantic-scale tasks. Omni IIE Bench provides critical diagnostic tools and insights for the development of next-generation, more reliable, and stable IIE models.

cs.CV↗

arXiv2026

Coupling Experts and Routers in Mixture-of-Experts via an Auxiliary Loss

Mixture-of-Experts (MoE) models lack explicit constraints to ensure the router's decisions align well with the experts' capabilities, which ultimately limits model performance. To address this, we propose expert-router coupling (ERC) loss, a lightweight auxiliary loss that tightly couples the router's decisions with expert capabilities. Our approach treats each expert's router embedding as a proxy token for the tokens assigned to that expert, and feeds perturbed router embeddings through the experts to obtain intermediate activations. The ERC loss enforces two constraints on these activations: (1) Each expert must exhibit higher activation for its own proxy token than for the proxy tokens of any other expert. (2) Each proxy token must elicit stronger activation from its corresponding expert than from any other expert. These constraints jointly ensure that each router embedding faithfully represents its corresponding expert's capability, while each expert specializes in processing the tokens actually routed to it. The ERC loss is computationally efficient, operating only on $n^2$ activations, where $n$ is the number of experts. This represents a fixed cost independent of batch size, unlike prior coupling methods that scale with the number of tokens (often millions per batch). Through pre-training MoE-LLMs ranging from 3B to 15B parameters and extensive analysis on trillions of tokens, we demonstrate the effectiveness of the ERC loss. Moreover, the ERC loss offers flexible control and quantitative tracking of expert specialization levels during training, providing valuable insights into MoEs.

cs.CL↗

arXiv2026

Beyond Closed-Pool Video Retrieval: A Benchmark and Agent Framework for Real-World Video Search and Moment Localization

Traditional video retrieval benchmarks focus on matching precise descriptions to closed video pools, failing to reflect real-world searches characterized by fuzzy, multi-dimensional memories on the open web. We present \textbf{RVMS-Bench}, a comprehensive system for evaluating real-world video memory search. It consists of \textbf{1,440 samples} spanning \textbf{20 diverse categories} and \textbf{four duration groups}, sourced from \textbf{real-world open-web videos}. RVMS-Bench utilizes a hierarchical description framework encompassing \textbf{Global Impression, Key Moment, Temporal Context, and Auditory Memory} to mimic realistic multi-dimensional search cues, with all samples strictly verified via a human-in-the-loop protocol. We further propose \textbf{RACLO}, an agentic framework that employs abductive reasoning to simulate the human ``Recall-Search-Verify'' cognitive process, effectively addressing the challenge of searching for videos via fuzzy memories in the real world. Experiments reveal that existing MLLMs still demonstrate insufficient capabilities in real-world Video Retrieval and Moment Localization based on fuzzy memories. We believe this work will facilitate the advancement of video retrieval robustness in real-world unstructured scenarios.

cs.CV↗

arXiv2026

Dynamic Passivity Multipliers for Plug-and-Play Stability Certificates of Converter-Dominated Grids

Ensuring small-signal stability in power systems with a high share of inverter-based resources (IBRs) is hampered by two factors: (i) device and network parameters are often uncertain or completely unknown, and (ii) brute-force enumeration of all topologies is computationally intractable. These challenges motivate plug-and-play (PnP) certificates that verify stability locally yet hold globally. Passivity is an attractive property because it guarantees stability under feedback and network interconnections; however, strict passivity rarely holds for practical controllers such as Grid Forming Inverters (GFMs) employing P-Q droop. This paper extends the passivity condition by constructing a dynamic, frequency-dependent multiplier that enables PnP stability certification of each component based solely on its admittance, without requiring any modification to the controller design. The multiplier is parameterised as a linear filter whose coefficients are tuned under a passivity goal. Numerical results for practical droop gains confirm the PnP rules, substantially enlarging the certified stability region while preserving the decentralised, model-agnostic nature of passivity-based PnP tests.

eess.SY↗

arXiv2025

Wealth or Stealth? The Camouflage Effect in Insider Trading

We consider a Kyle-type model where insider trading takes place among a potentially large population of liquidity traders and is subject to legal penalties. Insiders exploit the liquidity provided by the trading masses to "camouflage" their actions and balance expected wealth with the necessary stealth to avoid detection. Under a diverse spectrum of prosecution schemes, we establish the existence of equilibria for arbitrary population sizes and a unique limiting equilibrium. A convergence analysis determines the scale of insider trading by a stealth index $γ$, revealing that the equilibrium can be closely approximated by a simple limit due to diminished price informativeness. Empirical aspects are derived from two calibration experiments using non-overlapping data sets spanning from 1980 to 2018, which underline the indispensable role of a large population in insider trading models with legal risk, along with important implications for the incidence of stealth trading and the deterrent effect of legal enforcement.

econ.GN↗

arXiv2025

Superhedging under Proportional Transaction Costs in Continuous Time

We revisit the well-studied superhedging problem under proportional transaction costs in continuous time using the recently developed tools of set-valued stochastic analysis. By relying on a simple Black-Scholes-type market model for mid-prices and using continuous trading schemes, we define a dynamic family of superhedging sets in continuous time and express them in terms of set-valued integrals. We show that these sets, defined as subsets of Lebesgue spaces at different times, form a dynamic set-valued risk measure with multi-portfolio time-consistency. Finally, we transfer the problem formulation to a path-space setting and introduce approximate versions of superhedging sets that will involve relaxing the superhedging inequality, the superhedging probability, and the solvency requirement for the superhedging strategy with a predetermined error level. In this more technical framework, we are able to relate the approximate superhedging sets at different times by means of a set-valued Bellman's principle, which we believe will pave the way for a set-valued differential structure that characterizes the superhedging sets.

q-fin.RM↗

arXiv2025

Virtual Width Networks

We introduce Virtual Width Networks (VWN), a framework that delivers the benefits of wider representations without incurring the quadratic cost of increasing the hidden size. VWN decouples representational width from backbone width, expanding the embedding space while keeping backbone compute nearly constant. In our large-scale experiment, an 8-times expansion accelerates optimization by over 2 times for next-token and 3 times for next-2-token prediction. The advantage amplifies over training as both the loss gap grows and the convergence-speedup ratio increases, showing that VWN is not only token-efficient but also increasingly effective with scale. Moreover, we identify an approximately log-linear scaling relation between virtual width and loss reduction, offering an initial empirical basis and motivation for exploring virtual-width scaling as a new dimension of large-model efficiency.

cs.LG↗

arXiv2025

TFRank: Think-Free Reasoning Enables Practical Pointwise LLM Ranking

Reasoning-intensive ranking models built on Large Language Models (LLMs) have made notable progress. However, existing approaches often rely on large-scale LLMs and explicit Chain-of-Thought (CoT) reasoning, resulting in high computational cost and latency that limit real-world use. To address this, we propose \textbf{TFRank}, an efficient pointwise reasoning ranker based on small-scale LLMs. To improve ranking performance, TFRank effectively integrates CoT data, fine-grained score supervision, and multi-task training. Furthermore, it achieves an efficient ``\textbf{T}hink-\textbf{F}ree" reasoning capability by employing a ``think-mode switch'' and pointwise format constraints. Specifically, this allows the model to leverage explicit reasoning during training while delivering precise relevance scores for complex queries at inference without generating any reasoning chains. Experiments show that TFRank achieves performance comparable to models with four times more parameters on the BRIGHT benchmark and demonstrates strong competitiveness on the BEIR benchmark. Further analysis shows that TFRank achieves an effective balance between performance and efficiency, providing a practical solution for integrating advanced reasoning into real-world systems. Our code and data are released in the repository: https://github.com/JOHNNY-fans/TFRank.

cs.IR↗

arXiv2025

FLAME: Flexible and Lightweight Biometric Authentication Scheme in Malicious Environments

Privacy-preserving biometric authentication (PPBA) enables client authentication without revealing sensitive biometric data, addressing privacy and security concerns. Many studies have proposed efficient cryptographic solutions to this problem based on secure multi-party computation, typically assuming a semi-honest adversary model, where all parties follow the protocol but may try to learn additional information. However, this assumption often falls short in real-world scenarios, where adversaries may behave maliciously and actively deviate from the protocol. In this paper, we propose, implement, and evaluate $\sysname$, a \underline{F}lexible and \underline{L}ightweight biometric \underline{A}uthentication scheme designed for a \underline{M}alicious \underline{E}nvironment. By hybridizing lightweight secret-sharing-family primitives within two-party computation, $\sysname$ carefully designs a line of supporting protocols that incorporate integrity checks with rationally extra overhead. Additionally, $\sysname$ enables server-side authentication with various similarity metrics through a cross-metric-compatible design, enhancing flexibility and robustness without requiring any changes to the server-side process. A rigorous theoretical analysis validates the correctness, security, and efficiency of $\sysname$. Extensive experiments highlight $\sysname$'s superior efficiency, with a communication reduction by {$97.61\times \sim 110.13\times$} and a speedup of {$ 2.72\times \sim 2.82\times$ (resp. $ 6.58\times \sim 8.51\times$)} in a LAN (resp. WAN) environment, when compared to the state-of-the-art work.

cs.CR↗

arXiv2025

INT v.s. FP: A Comprehensive Study of Fine-Grained Low-bit Quantization Formats

Modern AI hardware, such as Nvidia's Blackwell architecture, is increasingly embracing low-precision floating-point (FP) formats to handle the pervasive activation outliers in Large Language Models (LLMs). Despite this industry trend, a unified comparison of FP and integer (INT) quantization across varying granularities has been missing, leaving algorithm and hardware co-design without clear guidance. This paper fills that gap by systematically investigating the trade-offs between FP and INT formats. We reveal a critical performance crossover: while FP excels in coarse-grained quantization, the comparison at fine-grained (block-wise) levels is more nuanced. Our comprehensive comparison demonstrates that for popular 8-bit fine-grained formats (e.g., MX with block size 32), MXINT8 is superior to its FP counterpart in both algorithmic accuracy and hardware efficiency. However, for 4-bit formats, FP (e.g., MXFP4, NVFP4) often holds an accuracy advantage , though we show that NVINT4 can surpass NVFP4 when outlier-mitigation techniques like Hadamard rotation are applied. We also introduce a symmetric clipping method that resolves gradient bias in fine-grained low-bit INT training, enabling nearly lossless performance for MXINT8 training. These findings challenge the current hardware trajectory, demonstrating that a one-size-fits-all FP approach is suboptimal and advocating that fine-grained INT formats, particularly MXINT8, offer a better balance of accuracy, power, and efficiency for future AI accelerators.

cs.LG↗

arXiv2025

Schrödinger bridge for generative AI: Soft-constrained formulation and convergence analysis

Generative AI can be framed as the problem of learning a model that maps simple reference measures into complex data distributions, and it has recently found a strong connection to the classical theory of the Schrödinger bridge problems (SBPs) due partly to their common nature of interpolating between prescribed marginals via entropy-regularized stochastic dynamics. However, the classical SBP enforces hard terminal constraints, which often leads to instability in practical implementations, especially in high-dimensional or data-scarce regimes. To address this challenge, we follow the idea of the so-called soft-constrained Schrödinger bridge problem (SCSBP), in which the terminal constraint is replaced by a general penalty function. This relaxation leads to a more flexible stochastic control formulation of McKean-Vlasov type. We establish the existence of optimal solutions for all penalty levels and prove that, as the penalty grows, both the controls and value functions converge to those of the classical SBP at a linear rate. Our analysis builds on Doob's h-transform representations, the stability results of Schrödinger potentials, Gamma-convergence, and a novel fixed-point argument that couples an optimization problem over the space of measures with an auxiliary entropic optimal transport problem. These results not only provide the first quantitative convergence guarantees for soft-constrained bridges but also shed light on how penalty regularization enables robust generative modeling, fine-tuning, and transfer learning.

cs.LG↗