arXiv subjects

Xiangyu Li

Publications and source records attributed to Xiangyu Li.

At least 91 records · Page 5Linked to original sources

arXiv2026

FinDeepForecast: A Live Multi-Agent System for Benchmarking Deep Research Agents in Financial Forecasting

Deep Research (DR) Agents powered by advanced Large Language Models (LLMs) have fundamentally shifted the paradigm for completing complex research tasks. Yet, a comprehensive and live evaluation of their forecasting performance on real-world, research-oriented tasks in high-stakes domains (e.g., finance) remains underexplored. We introduce FinDeepForecast, the first live, end-to-end multi-agent system for automatically evaluating DR agents by continuously generating research-oriented financial forecasting tasks. This system is equipped with a dual-track taxonomy, enabling the dynamic generation of recurrent and non-recurrent forecasting tasks at both corporate and macro levels. With this system, we generate FinDeepForecastBench, a weekly evaluation benchmark over a ten-week horizon, encompassing 8 global economies and 1,314 listed companies, and evaluate 13 representative methods. Extensive experiments show that, while DR agents consistently outperform strong baselines, their performance still falls short of genuine forward-looking financial reasoning. We expect the proposed FinDeepForecast system to consistently facilitate future advancements of DR agents in research-oriented financial forecasting tasks. The benchmark and leaderboard are publicly available on the OpenFinArena Platform.

cs.MA↗

arXiv2025

Reconfigurable Intelligent Surface-Enhanced Satellite Networks: Deployment Strategies, Key Capabilities, Practical Solutions, and Future Directions

Satellite networks promise wide-area 6G coverage but face two persistent barriers: blockage-induced service discontinuities and increasingly stringent spectrum coexistence across satellite layers and with terrestrial systems. Reconfigurable intelligent surfaces (RISs) act as low-power programmable apertures that redirect energy without the cost and power consumption of fully active arrays. We develop a deployment-first, operations-aware view of RIS-enabled satellite networking that treats RIS as both satellite/terminal antennas and inter-satellite or space-ground relays. We show that system-level gains are governed by two unifying mechanisms: connectivity restoration via virtual line-of-sight links that preserve connectivity under blockage and mobility, and angular selectivity that reshapes interference to enlarge spectrum reuse. We further discuss practical operation under high mobility, highlighting Delay-Doppler channel acquisition, predictive beam tracking, and control designs that budget overhead and latency, and summarize hardware considerations for reliable operation in space. Finally, we outline forward-looking opportunities in the generative artificial intelligence paradigm, multifunctional RIS architectures, ubiquitous satellite integrated sensing and communication, and sustainable satellite Internet-of-Things.

eess.SP↗

arXiv2025

APT-ClaritySet: A Large-Scale, High-Fidelity Labeled Dataset for APT Malware with Alias Normalization and Graph-Based Deduplication

Large-scale, standardized datasets for Advanced Persistent Threat (APT) research are scarce, and inconsistent actor aliases and redundant samples hinder reproducibility. This paper presents APT-ClaritySet and its construction pipeline that normalizes threat actor aliases (reconciling approximately 11.22\% of inconsistent names) and applies graph-feature deduplication -- reducing the subset of statically analyzable executables by 47.55\% while retaining behaviorally distinct variants. APT-ClaritySet comprises: (i) APT-ClaritySet-Full, the complete pre-deduplication collection with 34{,}363 malware samples attributed to 305 APT groups (2006 - early 2025); (ii) APT-ClaritySet-Unique, the deduplicated release with 25{,}923 unique samples spanning 303 groups and standardized attributions; and (iii) APT-ClaritySet-FuncReuse, a function-level resource that includes 324{,}538 function-reuse clusters (FRCs) enabling measurement of inter-/intra-group sharing, evolution, and tooling lineage. By releasing these components and detailing the alias normalization and scalable deduplication pipeline, this work provides a high-fidelity, reproducible foundation for quantitative studies of APT patterns, evolution, and attribution.

cs.CR↗

arXiv2025

Ambiguity-aware Truncated Flow Matching for Ambiguous Medical Image Segmentation

A simultaneous enhancement of accuracy and diversity of predictions remains a challenge in ambiguous medical image segmentation (AMIS) due to the inherent trade-offs. While truncated diffusion probabilistic models (TDPMs) hold strong potential with a paradigm optimization, existing TDPMs suffer from entangled accuracy and diversity of predictions with insufficient fidelity and plausibility. To address the aforementioned challenges, we propose Ambiguity-aware Truncated Flow Matching (ATFM), which introduces a novel inference paradigm and dedicated model components. Firstly, we propose Data-Hierarchical Inference, a redefinition of AMIS-specific inference paradigm, which enhances accuracy and diversity at data-distribution and data-sample level, respectively, for an effective disentanglement. Secondly, Gaussian Truncation Representation (GTR) is introduced to enhance both fidelity of predictions and reliability of truncation distribution, by explicitly modeling it as a Gaussian distribution at $T_{\text{trunc}}$ instead of using sampling-based approximations. Thirdly, Segmentation Flow Matching (SFM) is proposed to enhance the plausibility of diverse predictions by extending semantic-aware flow transformation in Flow Matching (FM). Comprehensive evaluations on LIDC and ISIC3 datasets demonstrate that ATFM outperforms SOTA methods and simultaneously achieves a more efficient inference. ATFM improves GED and HM-IoU by up to $12\%$ and $7.3\%$ compared to advanced methods.

cs.CV↗

arXiv2025

PAVE: An End-to-End Dataset for Production Autonomous Vehicle Evaluation

Most existing autonomous-driving datasets (e.g., KITTI, nuScenes, and the Waymo Perception Dataset), collected by human-driving mode or unidentified driving mode, can only serve as early training for the perception and prediction of autonomous vehicles (AVs). To evaluate the real behavioral safety of AVs controlled in the black box, we present the first end-to-end benchmark dataset collected entirely by autonomous-driving mode in the real world. This dataset contains over 100 hours of naturalistic data from multiple production autonomous-driving vehicle models in the market. We segment the original data into 32,727 key frames, each consisting of four synchronized camera images and high-precision GNSS/IMU data (0.8 cm localization accuracy). For each key frame, 20 Hz vehicle trajectories spanning the past 6 s and future 5 s are provided, along with detailed 2D annotations of surrounding vehicles, pedestrians, traffic lights, and traffic signs. These key frames have rich scenario-level attributes, including driver intent, area type (covering highways, urban roads, and residential areas), lighting (day, night, or dusk), weather (clear or rain), road surface (paved or unpaved), traffic and vulnerable road users (VRU) density, traffic lights, and traffic signs (warning, prohibition, and indication). To evaluate the safety of AVs, we employ an end-to-end motion planning model that predicts vehicle trajectories with an Average Displacement Error (ADE) of 1.4 m on autonomous-driving frames. The dataset continues to expand by over 10 hours of new data weekly, thereby providing a sustainable foundation for research on AV driving behavior analysis and safety evaluation. The PAVE dataset is publicly available at https://hkustgz-my.sharepoint.com/:f:/g/personal/kema_hkust-gz_edu_cn/IgDXyoHKfdGnSZ3JbbidjduMAXxs-Z3NXzm005A_Ix9tr0Q?e=9HReCu.

cs.CV↗

arXiv2025

DuoZone: A User-Centric, LLM-Guided Mixed-Initiative XR Window Management System

Mixed reality (XR) environments offer vast spatial possibilities, but current window management systems require users to manually place, resize, and organize multiple applications across large 3D spaces. This creates cognitive and interaction burdens that limit productivity. We introduce DuoZone, a mixed-initiative XR window management system that combines user-defined spatial layouts with LLM-guided automation. DuoZone separates window management into two complementary zones. The Recommendation Zone enables fast setup by providing spatial layout templates and automatically recommending relevant applications based on user tasks and high-level goals expressed through voice or text. The Arrangement Zone supports precise refinement through direct manipulation, allowing users to adjust windows using natural spatial actions such as dragging, resizing, and snapping. Through this dual-zone approach, DuoZone promotes efficient organization while reducing user cognitive load. We conducted a user study comparing DuoZone with a baseline manual XR window manager. Results show that DuoZone improves task completion speed, reduces mental effort, and increases sense of control when working with multiple applications in XR. We discuss design implications for future mixed-initiative systems and outline opportunities for integrating adaptive, goal-aware intelligence into spatial computing workflows.

cs.HC↗

arXiv2025

Downlink Performance of Cell-Free Massive MIMO for LEO Satellite Mega-Constellation

Low-earth orbit (LEO) satellite communication (SatCom) has emerged as a promising technology to improve wireless connectivity in global areas. Cell-free massive multiple-input multiple-output (CF-mMIMO), an architecture proposed for next-generation networks, has yet to be fully explored for LEO satellites. In this paper, we investigate the downlink performance of a CF-mMIMO LEO SatCom network, where multiple satellite access points (SAPs) simultaneously serve the corresponding ground user terminals (UTs). Using tools from stochastic geometry, we model the locations of SAPs and UTs on surfaces of concentric spheres using Poisson point processes (PPPs) and present expressions on transmit and received signals, signal-to-interference-plus-noise ratio (SINR). Then, we derive the coverage probabilities in fading scenarios, considering significant system parameters such as the Nakagami fading parameter, the number of UTs, the number of SAPs, the orbital altitude, and the service range affected by the dome angle. Finally, the analytical model is verified by extensive Monte Carlo simulations. Simulation results indicate that stronger line-of-sight (LoS) effects and a more comprehensive service range of the UT result in a higher coverage probability, despite the presence of multi-user interference (MUI). Moreover, we found that there exist optimal numbers of UTs that maximize system capacity for different orbital altitudes and dome angles, providing valuable insights for system design.

eess.SP↗

arXiv2025

CDI-DTI: A Strong Cross-domain Interpretable Drug-Target Interaction Prediction Framework Based on Multi-Strategy Fusion

Accurate prediction of drug-target interactions (DTI) is pivotal for drug discovery, yet existing methods often fail to address challenges like cross-domain generalization, cold-start prediction, and interpretability. In this work, we propose CDI-DTI, a novel cross-domain interpretable framework for DTI prediction, designed to overcome these limitations. By integrating multi-modal features-textual, structural, and functional-through a multi-strategy fusion approach, CDI-DTI ensures robust performance across different domains and in cold-start scenarios. A multi-source cross-attention mechanism is introduced to align and fuse features early, while a bidirectional cross-attention layer captures fine-grained intra-modal drug-target interactions. To enhance model interpretability, we incorporate Gram Loss for feature alignment and a deep orthogonal fusion module to eliminate redundancy. Experimental results on several benchmark datasets demonstrate that CDI-DTI significantly outperforms existing methods, particularly in cross-domain and cold-start tasks, while maintaining high interpretability for practical applications in drug-target interaction prediction.

cs.MM↗

arXiv2025

RAD: Training an End-to-End Driving Policy via Large-Scale 3DGS-based Reinforcement Learning

Existing end-to-end autonomous driving (AD) algorithms typically follow the Imitation Learning (IL) paradigm, which faces challenges such as causal confusion and an open-loop gap. In this work, we propose RAD, a 3DGS-based closed-loop Reinforcement Learning (RL) framework for end-to-end Autonomous Driving. By leveraging 3DGS techniques, we construct a photorealistic digital replica of the real physical world, enabling the AD policy to extensively explore the state space and learn to handle out-of-distribution scenarios through large-scale trial and error. To enhance safety, we design specialized rewards to guide the policy in effectively responding to safety-critical events and understanding real-world causal relationships. To better align with human driving behavior, we incorporate IL into RL training as a regularization term. We introduce a closed-loop evaluation benchmark consisting of diverse, previously unseen 3DGS environments. Compared to IL-based methods, RAD achieves stronger performance in most closed-loop metrics, particularly exhibiting a 3x lower collision rate. Abundant closed-loop results are presented in the supplementary material. Code is available at https://github.com/hustvl/RAD for facilitating future research.

cs.CV↗

arXiv2025

M3ST-DTI: A multi-task learning model for drug-target interactions based on multi-modal features and multi-stage alignment

Accurate prediction of drug-target interactions (DTI) is pivotal in drug discovery. However, existing approaches often fail to capture deep intra-modal feature interactions or achieve effective cross-modal alignment, limiting predictive performance and generalization. To address these challenges, we propose M3ST-DTI, a multi-task learning model that enables multi-stage integration and alignment of multi modal features for DTI prediction. M3ST-DTI incorporates three types of features-textual, structural, and functional and enhances intra-modal representations using self-attention mechanisms and a hybrid pooling graph attention module. For early-stage feature alignment and fusion, the model in tegrates MCA with Gram loss as a structural constraint. In the later stage, a BCA module captures fine-grained interactions between drugs and targets within each modality, while a deep orthogonal fusion module mitigates feature redundancy.Extensive evaluations on benchmark datasets demonstrate that M3ST-DTI consistently outperforms state-of-the art methods across diverse metrics

cs.MM↗

arXiv2025

Profit Mirage: Revisiting Information Leakage in LLM-based Financial Agents

LLM-based financial agents have attracted widespread excitement for their ability to trade like human experts. However, most systems exhibit a "profit mirage": dazzling back-tested returns evaporate once the model's knowledge window ends, because of the inherent information leakage in LLMs. In this paper, we systematically quantify this leakage issue across four dimensions and release FinLake-Bench, a leakage-robust evaluation benchmark. Furthermore, to mitigate this issue, we introduce FactFin, a framework that applies counterfactual perturbations to compel LLM-based agents to learn causal drivers instead of memorized outcomes. FactFin integrates four core components: Strategy Code Generator, Retrieval-Augmented Generation, Monte Carlo Tree Search, and Counterfactual Simulator. Extensive experiments show that our method surpasses all baselines in out-of-sample generalization, delivering superior risk-adjusted performance.

cs.AI↗

arXiv2025

Disentangling peri-urban river hypoxia

Episodes of low dissolved oxygen concentration--hypoxia--threaten the functioning of and the services provided by aquatic ecosystems, particularly those of urban rivers. Here, we disentangle oxygen-related processes in the highly modified Elbe River flowing through the major German city of Hamburg, where low oxygen levels are frequently observed. We use a process-based biochemical model that describes particulate and dissolved organic matter, micro-algae, their pathogens, and the key reactions that produce or consume oxygen: photosynthesis, re-aeration, respiration, mineralization, and nitrification. The model analysis reveals pronounced spatial variability in the relative importance of these processes. Photosynthesis and respiration are more prominent upstream of the city, while mineralization, nitrification, and re-aeration prevail downstream. The city, characterized by rapid changes in bathymetry, marks a transitional area: pathogen-related micro-algal lysis may increase organic material, explaining the shift towards heterotrophic processes downstream. As the primary driver of seasonal changes, the model analysis reveals a differential temperature sensitivity of biochemical rates. These results may be extrapolated to other urban rivers, and also provide valuable information for estuarine water quality management.

q-bio.OT↗

arXiv2025

QuantAgents: Towards Multi-agent Financial System via Simulated Trading

In this paper, our objective is to develop a multi-agent financial system that incorporates simulated trading, a technique extensively utilized by financial professionals. While current LLM-based agent models demonstrate competitive performance, they still exhibit significant deviations from real-world fund companies. A critical distinction lies in the agents' reliance on ``post-reflection'', particularly in response to adverse outcomes, but lack a distinctly human capability: long-term prediction of future trends. Therefore, we introduce QuantAgents, a multi-agent system integrating simulated trading, to comprehensively evaluate various investment strategies and market scenarios without assuming actual risks. Specifically, QuantAgents comprises four agents: a simulated trading analyst, a risk control analyst, a market news analyst, and a manager, who collaborate through several meetings. Moreover, our system incentivizes agents to receive feedback on two fronts: performance in real-world markets and predictive accuracy in simulated trading. Extensive experiments demonstrate that our framework excels across all metrics, yielding an overall return of nearly 300% over the three years (https://quantagents.github.io/).

cs.AI↗

arXiv2025

TimeExpert: Boosting Long Time Series Forecasting with Temporal Mix of Experts

Transformer-based architectures dominate time series modeling by enabling global attention over all timestamps, yet their rigid 'one-size-fits-all' context aggregation fails to address two critical challenges in real-world data: (1) inherent lag effects, where the relevance of historical timestamps to a query varies dynamically; (2) anomalous segments, which introduce noisy signals that degrade forecasting accuracy. To resolve these problems, we propose the Temporal Mix of Experts (TMOE), a novel attention-level mechanism that reimagines key-value (K-V) pairs as local experts (each specialized in a distinct temporal context) and performs adaptive expert selection for each query via localized filtering of irrelevant timestamps. Complementing this local adaptation, a shared global expert preserves the Transformer's strength in capturing long-range dependencies. We then replace the vanilla attention mechanism in popular time-series Transformer frameworks (i.e., PatchTST and Timer) with TMOE, without extra structural modifications, yielding our specific version TimeExpert and general version TimeExpert-G. Extensive experiments on seven real-world long-term forecasting benchmarks demonstrate that TimeExpert and TimeExpert-G outperform state-of-the-art methods. Code is available at https://github.com/xwmaxwma/TimeExpert.

cs.LG↗

arXiv2025

Quantum Algorithm for Subcellular Multiscale Reaction-Diffusion Systems

Computational modeling of cellular systems, where reactants are governed by biochemical equations and physical representations, requires extensive classical computing resources. These limitations significantly constrain the system size and spatiotemporal scales of simulations. A key challenge lies in the "curse of dimensionality", where the number of possible reaction terms grows exponentially with the number of species, and the computation of reaction rates involving many-body interactions becomes intractable in polynomial time on classical computers. In this work, we introduce a quantum algorithmic framework designed to overcome these challenges, leveraging the architecture of quantum computing to simultaneously compute reaction rates and track the spatiotemporal dynamical evolutions of subcellular systems. We generalize the reaction-diffusion equation (RDE) for multiscale systems with arbitrary species count, encompassing higher-order interactions. Our approach achieves two principal quantum advantages: (i) an exponential quantum speedup in reaction-rate computation, contingent on the efficient preparation of polynomially accurate ground states on a quantum computer, and (ii)) a quadratic scaling in spatial grid points and polynomial scaling in the number of species for solving nonlinear RDEs, contrasting sharply with classical methods that scale exponentially with the system's degrees of freedom. To our knowledge, this represents the first efficient quantum algorithm for solving multiscale reaction-diffusion systems. This framework opens the door to simulations of biologically relevant subcellular processes across previously inaccessible spatial and temporal scales, with profound implications for computational biology, soft matter physics, and biophysical modeling.

quant-ph↗

arXiv2025

T2R-bench: A Benchmark for Generating Article-Level Reports from Real World Industrial Tables

Extensive research has been conducted to explore the capabilities of large language models (LLMs) in table reasoning. However, the essential task of transforming tables information into reports remains a significant challenge for industrial applications. This task is plagued by two critical issues: 1) the complexity and diversity of tables lead to suboptimal reasoning outcomes; and 2) existing table benchmarks lack the capacity to adequately assess the practical application of this task. To fill this gap, we propose the table-to-report task and construct a bilingual benchmark named T2R-bench, where the key information flow from the tables to the reports for this task. The benchmark comprises 457 industrial tables, all derived from real-world scenarios and encompassing 19 industry domains as well as 4 types of industrial tables. Furthermore, we propose an evaluation criteria to fairly measure the quality of report generation. The experiments on 25 widely-used LLMs reveal that even state-of-the-art models like Deepseek-R1 only achieves performance with 62.71 overall score, indicating that LLMs still have room for improvement on T2R-bench.

cs.CL↗

arXiv2025

Identifying Fixed Points in the Three-Body Problem Using a High-Order Transfer Map

Periodic orbits (POs) play a central role in the circular restricted three-body problem (CRTBP). This paper introduces a method to search for POs by identifying single- and multiple-revolution fixed points in chosen Poincare maps that describe the CRTBP dynamics, with a theoretical capability to detect all fixed points across arbitrary revolution counts exhaustively.First, high-order transfer maps (HOTMs), represented as polynomials, are constructed within the differential algebra (DA) framework for both planar and spatial CRTBP to map states between successive Poincare section crossings, with the Jacobi constant used to reduce the number of independent variables. Next, an automatic domain splitting (ADS) strategy is employed to generate subdomains, preserving HOTM accuracy, with an integrated feasibility estimation to reduce ADS's computation burden.Then, a two-stage HOTM-based polynomial optimization framework is introduced, first identifying combinable subdomain sequences and then refining the fixed point solutions. Finally, the method is applied to the Earth-Moon CRTBP, identifying POs up to nine revolutions in the planar case and four in the spatial case. Known families such as distant retrograde orbits (DROs) and Lyapunov orbits are recovered, along with a previously undocumented family that exhibits a hybrid character between DROs and Lyapunov orbits.

astro-ph.EP↗

arXiv2025

Boosting Micro-Expression Analysis via Prior-Guided Video-Level Regression

Micro-expressions (MEs) are involuntary, low-intensity, and short-duration facial expressions that often reveal an individual's genuine thoughts and emotions. Most existing ME analysis methods rely on window-level classification with fixed window sizes and hard decisions, which limits their ability to capture the complex temporal dynamics of MEs. Although recent approaches have adopted video-level regression frameworks to address some of these challenges, interval decoding still depends on manually predefined, window-based methods, leaving the issue only partially mitigated. In this paper, we propose a prior-guided video-level regression method for ME analysis. We introduce a scalable interval selection strategy that comprehensively considers the temporal evolution, duration, and class distribution characteristics of MEs, enabling precise spotting of the onset, apex, and offset phases. In addition, we introduce a synergistic optimization framework, in which the spotting and recognition tasks share parameters except for the classification heads. This fully exploits complementary information, makes more efficient use of limited data, and enhances the model's capability. Extensive experiments on multiple benchmark datasets demonstrate the state-of-the-art performance of our method, with an STRS of 0.0562 on CAS(ME)$^3$ and 0.2000 on SAMMLV. The code is available at https://github.com/zizheng-guo/BoostingVRME.

cs.CV↗