Searcharxiv⌕ Search

arXiv subjects

Xinyu Wang

Publications and source records attributed to Xinyu Wang.

At least 145 records · Page 8Linked to original sources
arXiv2026

ProxyWar: Dynamic Assessment of LLM Code Generation in Game Arenas

Large language models (LLMs) have revolutionized automated code generation, yet the evaluation of their real-world effectiveness remains limited by static benchmarks and simplistic metrics. We present ProxyWar, a novel framework that systematically assesses code generation quality by embedding LLM-generated agents within diverse, competitive game environments. Unlike existing approaches, ProxyWar evaluates not only functional correctness but also the operational characteristics of generated programs, combining automated testing, iterative code repair, and multi-agent tournaments to provide a holistic view of program behavior. Applied to a range of state-of-the-art coders and games, our approach uncovers notable discrepancies between benchmark scores and actual performance in dynamic settings, revealing overlooked limitations and opportunities for improvement. These findings highlight the need for richer, competition-based evaluation of code generation. Looking forward, ProxyWar lays a foundation for research into LLM-driven algorithm discovery, adaptive problem solving, and the study of practical efficiency and robustness, including the potential for models to outperform hand-crafted agents. The project is available at https://github.com/xinke-wang/ProxyWar.

cs.SE↗
arXiv2026

Modality-Tailored Age of Information for Multimodal Data in Edge Computing Systems

As Internet of Things (IoT) systems scale and device heterogeneity grows, multimodal data have become ubiquitous. Meanwhile, evaluating the freshness of multimodal data is essential, as stale updates would delay task execution, degrade decision accuracy, and undermine safety in latency-sensitive services. However, existing freshness metrics such as Age of Information (AoI) are not suitable for multimodal data, as they do not capture modality-specific characteristics. In this paper, we propose a metric, namely, Modality-Tailored Age of Information (MAoI), to provide a unified and decision-relevant evaluation of freshness for resource management and policy optimization for multimodal data. This metric integrates modality-specific semantic and temporal characteristics, reflecting both age evolution and content importance for multimodal data in multi-access edge computing (MEC) systems. Then, the closed-form expression of the average MAoI is derived, and an MAoI minimization problem is formulated, where sampling intervals and offloading decisions are optimized with practical energy constraints. To effectively solve this problem, a Joint Sampling Offloading Optimization (JSO) algorithm is proposed to jointly optimize the sampling intervals and offloading decisions. It is a block coordinate descent-based algorithm where an optimal sampling-interval subalgorithm is used to update the sampling intervals, and an interference-aware best-response offloading subalgorithm is proposed to update the offloading decisions alternately. Finally, a comprehensive simulation is performed, confirming that the MAoI metric effectively quantifies multimodal freshness compared to traditional AoI, and the JSO algorithm significantly minimizes the average MAoI compared to state-of-the-art algorithms.

cs.NI↗
arXiv2026

Design of a 60.8 K superconducting hydride LiMgZr2H12 at ambient pressure via Lithium doping

High-pressure hydrogen-rich compounds have long been regarded as promising room-temperature superconductor candidates; however, their practical applications are limited by their reliance on extreme compression. This study explores hydrogen-rich superconductors that may be stable at ambient pressures. Inspired by recent investigations of the MgZrH2n family, the LiMgZr2H12 structure with a Pmmm symmetry was constructed, and its thermodynamic, mechanical, and dynamical stability were evaluated using first-principles calculations. Electron-phonon coupling (EPC) analysis suggests that LiMgZr2H12 reaches a superconducting critical temperature (Tc) of 60.8 K at ambient pressure. Compared with MgZrH6, Li doping significantly increases the contribution of hydrogen atoms to the electron density of states near the Fermi level (EF) and enhances the EPC constant of the LiMgZr2H12 structure. LiMgZr2H12 exhibits a superconducting figure of merit of 1.56, which is significantly greater than that of MgZrH6, demonstrating its outstanding potential for practical applications. This work guides ambient-pressure design of high-Tc hydrides.

cond-mat.supr-con↗
arXiv2026

WFR-MFM: One-Step Inference for Dynamic Unbalanced Optimal Transport

Reconstructing dynamical evolution from limited observations is a fundamental challenge in single-cell biology, where dynamic unbalanced optimal transport provides a principled framework for modeling coupled transport and mass variation. However, existing approaches rely on trajectory simulation at inference time, making inference a key bottleneck for scalable applications. In this work, we propose a mean-flow framework for unbalanced flow matching that summarizes both transport and mass-growth dynamics over arbitrary time intervals using mean velocity and mass-growth fields, enabling fast one-step generation without trajectory simulation. To solve dynamic unbalanced optimal transport under the Wasserstein-Fisher-Rao geometry, we further build on this framework to develop Wasserstein-Fisher-Rao Mean Flow Matching (WFR-MFM). Across synthetic and real single-cell RNA sequencing datasets, WFR-MFM achieves orders-of-magnitude faster inference than a range of existing baselines while maintaining high predictive accuracy, and enables efficient perturbation response prediction on large synthetic datasets with thousands of conditions.

cs.LG↗
arXiv2026

Efficient Multimodal Planning Agent for Visual Question-Answering

Visual Question-Answering (VQA) is a challenging multimodal task that requires integrating visual and textual information to generate accurate responses. While multimodal Retrieval-Augmented Generation (mRAG) has shown promise in enhancing VQA systems by providing more evidence on both image and text sides, the default procedure that addresses VQA queries, especially the knowledge-intensive ones, often relies on multi-stage pipelines of mRAG with inherent dependencies. To mitigate the inefficiency limitations while maintaining VQA task performance, this paper proposes a method that trains a multimodal planning agent, dynamically decomposing the mRAG pipeline to solve the VQA task. Our method optimizes the trade-off between efficiency and effectiveness by training the agent to intelligently determine the necessity of each mRAG step. In our experiments, the agent can help reduce redundant computations, cutting search time by over 60\% compared to existing methods and decreasing costly tool calls. Meanwhile, experiments demonstrate that our method outperforms all baselines, including a Deep Research agent and a carefully designed prompt-based method, on average over six various datasets. Code will be released.

cs.CL↗
arXiv2026

HiCache: A Plug-in Scaled-Hermite Upgrade for Taylor-Style Cache-then-Forecast Diffusion Acceleration

Diffusion models have achieved remarkable success in content generation but often incur prohibitive computational costs due to iterative sampling. Recent feature caching methods accelerate inference via temporal extrapolation, yet can suffer quality degradation from inaccurate modeling of the complex dynamics of feature evolution. We propose HiCache (Hermite Polynomial-based Feature Cache), a training-free acceleration framework that improves feature prediction by aligning mathematical tools with empirical properties. Our key insight is that feature-derivative approximations in diffusion Transformers exhibit multivariate Gaussian characteristics, motivating the use of Hermite polynomials as a potentially optimal basis for Gaussian-correlated processes. We further introduce a dual-scaling mechanism that ensures numerical stability while preserving predictive accuracy, and is also effective when applied standalone or integrated with TaylorSeer. Extensive experiments demonstrate HiCache's superiority, achieving 5.55x speedup on FLUX.1-dev while matching or exceeding baseline quality, and maintaining strong performance across text-to-image, video generation, and super-resolution tasks. Moreover, HiCache can be naturally added to previous caching methods to enhance their performance, e.g., improving ClusCa from 0.9480 to 0.9840 in terms of image rewards. Code: https://github.com/fenglang918/HiCache

cs.CV↗
arXiv2026

Evolutionary Dynamics of Reputation-Based Voluntary Prisoner's Dilemma Games

Cooperation underlies many natural and artificial systems. While voluntary participation can sustain cooperation without informational assumptions, real interactions are rarely anonymous, leaving the joint effects of participation and reputation insufficiently understood. We propose a reputation-based voluntary Prisoner's Dilemma in which agents incur a monitoring cost to inspect opponents and decide whether to exit an interaction for a fixed incentive to avoid exploitation or to default to cooperation or defection. We show that reputation-conditioned exit generates multiple coexistence pathways that sustain cooperation across population structures. In well-mixed populations, cooperation persists through stable mixed coexistence, whereas in structured populations, exit-incentive-dependent regimes emerge, including local cyclic dominance and persistent oscillations. Together, these results extend voluntary participation frameworks and underscore the role of exit-incentive design in cooperative multi-agent systems.

physics.soc-ph↗
arXiv2026

Geo-Registration of Terrestrial LiDAR Point Clouds with Satellite Images without GNSS

Accurate geo-registration of LiDAR point clouds remains a significant challenge in urban environments where Global Navigation Satellite System (GNSS) signals are denied or degraded. Existing methods typically rely on real-time GNSS and Inertial Measurement Unit (IMU) data, which require pre-calibration and assume stable signals. However, this assumption often fails in dense cities, resulting in localization errors. To address this, we propose a structured post-hoc geo-registration method that accurately aligns LiDAR point clouds with satellite images. The proposed approach targets point cloud datasets where reliable GNSS information is unavailable or degraded, enabling city-scale geo-registration as a post-processing solution. Our method uses a pre-trained Point Transformer to segment road points, then extracts road skeletons and intersections from the point cloud and the satellite image. Global alignment is achieved through rigid transformation using corresponding intersection points, followed by local non-rigid refinement with radial basis function (RBF) interpolation. Elevation discrepancies are corrected using terrain data from the Shuttle Radar Topography Mission (SRTM). To evaluate geo-registration accuracy, we measure the absolute distances between the roads extracted from the two modalities. Our method is validated on the KITTI benchmark and a newly collected dataset of Perth, Western Australia. On KITTI, our method achieves a mean planimetric alignment error of 0.69m, corresponding to a 50% reduction in global geo-registration bias compared to the raw KITTI annotations. On Perth dataset, it achieves a mean planimetric error of 2.17m from GNSS values extracted from Google Maps, corresponding to 57.4% improvement over rigid alignment. Elevation correlation factor improved by 30.5% (KITTI) and 55.8% (Perth).

cs.CV↗
arXiv2026

Controllable Video Generation: A Survey

With the rapid development of AI-generated content (AIGC), video generation has emerged as one of its most dynamic and impactful subfields. In particular, the advancement of video generation foundation models has led to growing demand for controllable video generation methods that can more accurately reflect user intent. Most existing foundation models are designed for text-to-video generation, where text prompts alone are often insufficient to express complex, multi-modal, and fine-grained user requirements. This limitation makes it challenging for users to generate videos with precise control using current models. To address this issue, recent research has explored the integration of additional non-textual conditions, such as camera motion, depth maps, and human pose, to extend pretrained video generation models and enable more controllable video synthesis. These approaches aim to enhance the flexibility and practical applicability of AIGC-driven video generation systems. In this survey, we provide a systematic review of controllable video generation, covering both theoretical foundations and recent advances in the field. We begin by introducing the key concepts and commonly used open-source video generation models. We then focus on control mechanisms in video diffusion models, analyzing how different types of conditions can be incorporated into the denoising process to guide generation. Finally, we categorize existing methods based on the types of control signals they leverage, including single-condition generation, multi-condition generation, and universal controllable generation. For a complete list of the literature on controllable video generation reviewed, please visit our curated repository at https://github.com/mayuelala/Awesome-Controllable-Video-Generation.

cs.GR↗
arXiv2026

Wasserstein geometry of nonnegative measures on finite Markov chains I: Gradient flow

We investigate a Benamou--Brenier type transportation metric for nonnegative measures on a finite reversible Markov chain, which endows the space of measures with a Riemannian structure. Using this geometric framework, we identify a generalized heat equation with source as the gradient flow of the discrete entropy. Moreover, by means of a local Łojasiewicz inequality, we prove exponential convergence of the flow to a unique equilibrium. Our results clarify the role of the Benamou--Brenier formulation in discrete optimal transport for nonnegative measures and provide a coherent geometric interpretation of generalized diffusion equations with source terms.

math.AP↗
arXiv2026

Wasserstein geometry of nonnegative measures on finite Markov chains II: Geodesic and duality formulae

In this paper, we investigate the geodesic structure and the associated Kantorovich-type duality for a Benamou-Brenier-type transportation metric defined on the space of nonnegative measures over a finite reversible Markov chain. The metric is introduced through a dynamic formulation that combines transport and source costs along solutions of a nonconservative continuity equation, where mass variation is constrained to occur along a fixed strictly positive reference direction. We show that geodesics associated with this metric exhibit a non-locality property: almost every time, they are supported on the whole state space, independently of the choice of endpoints. Moreover, along optimal curves, the source term displays a characteristic temporal profile, with mass creation occurring at early times and subsequent decay as the curve approaches the target measure. As an application of this property, we compare our metric with the shift-transport distance and prove that the latter is always bounded above by our metric. Finally, we establish a Kantorovich-type duality formula in terms of Hamilton-Jacobi subsolutions, which provides a characterization of the metric and highlights the role of the momentum associated with geodesic curves.

math.AP↗
arXiv2026

CymbaDiff: Structured Spatial Diffusion for Sketch-based 3D Semantic Urban Scene Generation

Outdoor 3D semantic scene generation produces realistic and semantically rich environments for applications such as urban simulation and autonomous driving. However, advances in this direction are constrained by the absence of publicly available, well-annotated datasets. We introduce SketchSem3D, the first large-scale benchmark for generating 3D outdoor semantic scenes from abstract freehand sketches and pseudo-labeled annotations of satellite images. SketchSem3D includes two subsets, Sketch-based SemanticKITTI and Sketch-based KITTI-360 (containing LiDAR voxels along with their corresponding sketches and annotated satellite images), to enable standardized, rigorous, and diverse evaluations. We also propose Cylinder Mamba Diffusion (CymbaDiff) that significantly enhances spatial coherence in outdoor 3D scene generation. CymbaDiff imposes structured spatial ordering, explicitly captures cylindrical continuity and vertical hierarchy, and preserves both physical neighborhood relationships and global context within the generated scenes. Extensive experiments on SketchSem3D demonstrate that CymbaDiff achieves superior semantic consistency, spatial realism, and cross-dataset generalization. The code and dataset will be available at https://github.com/Lillian-research-hub/CymbaDiff

cs.CV↗
arXiv2026

Semiparametric inference for inequality measures under nonignorable nonresponse using callback data

This paper develops semiparametric methods for estimation and inference of widely used inequality measures when survey data are subject to nonignorable nonresponse, a challenging setting in which response probabilities depend on the unobserved outcomes. Such nonresponse mechanisms are common in household surveys and invalidate standard inference procedures due to selection bias and lack of population representativeness. We address this problem by exploiting callback data from repeated contact attempts and adopting a semiparametric model that leaves the outcome distribution unspecified. We construct semiparametric full-likelihood estimators for the underlying distribution and the associated inequality measures, and establish their large-sample properties for a broad class of functionals, including quantiles, the Theil index, and the Gini index. Explicit asymptotic variance expressions are derived, enabling valid Wald-type inference under nonignorable nonresponse. To facilitate implementation, we propose a stable and computationally convenient expectation-maximization algorithm, whose steps either admit closed-form expressions or reduce to fitting a standard logistic regression model. Simulation studies demonstrate that the proposed procedures effectively correct nonresponse bias and achieve near-benchmark efficiency. An application to Consumer Expenditure Survey data illustrates the practical gains from incorporating callback information when making inference on inequality measures.

econ.EM↗
arXiv2026

Style-constrained inverse design of microstructures with tailored mechanical properties using unconditional diffusion models

Deep generative models, particularly denoising diffusion models, have achieved remarkable success in high-fidelity generation of architected microstructures with desired properties and styles. Nevertheless, these recent methods typically rely on conditional training mechanisms and demand substantial computational effort to prepare the labeled training dataset, which makes them inflexible since any change in the governing equations or boundary conditions requires a complete retraining process. In this study, we propose a new inverse design framework that integrates unconditional denoising diffusion models with differentiable programming techniques for architected microstructure generation. Our approach eliminates the need for expensive labeled dataset preparation and retraining for different problem settings. By reinterpreting the noise input to the diffusion model as an optimizable design variable, we formulate the design task as an optimization problem over the noise input, enabling control over the reverse denoising trajectory to guide the generated microstructure toward the desired mechanical properties while preserving the stylistic constraints encoded in the training dataset. A unified differentiation pipeline via vector-Jacobian product concatenations is developed to enable end-to-end gradient evaluation through backpropagation. Several numerical examples, ranging from the design of microstructures with specified homogenized properties to those with targeted hyperelastic and elasto-plastic behaviors, showcase the effectiveness of the framework and its potential for advanced design tasks involving diverse performance and style requirements.

cs.CE↗
arXiv2026

SWE-Lego: Pushing the Limits of Supervised Fine-tuning for Software Issue Resolving

We present SWE-Lego, a supervised fine-tuning (SFT) recipe designed to achieve state-ofthe-art performance in software engineering (SWE) issue resolving. In contrast to prevalent methods that rely on complex training paradigms (e.g., mid-training, SFT, reinforcement learning, and their combinations), we explore how to push the limits of a lightweight SFT-only approach for SWE tasks. SWE-Lego comprises three core building blocks, with key findings summarized as follows: 1) the SWE-Lego dataset, a collection of 32k highquality task instances and 18k validated trajectories, combining real and synthetic data to complement each other in both quality and quantity; 2) a refined SFT procedure with error masking and a difficulty-based curriculum, which demonstrably improves action quality and overall performance. Empirical results show that with these two building bricks alone,the SFT can push SWE-Lego models to state-of-the-art performance among open-source models of comparable size on SWE-bench Verified: SWE-Lego-Qwen3-8B reaches 42.2%, and SWE-Lego-Qwen3-32B attains 52.6%. 3) We further evaluate and improve test-time scaling (TTS) built upon the SFT foundation. Based on a well-trained verifier, SWE-Lego models can be significantly boosted--for example, 42.2% to 49.6% and 52.6% to 58.8% under TTS@16 for the 8B and 32B models, respectively.

cs.SE↗
arXiv2026

Block-Diagonal LoRA for Eliminating Communication Overhead in Tensor Parallel LoRA Serving

When serving a single base LLM with several different LoRA adapters simultaneously, the adapters cannot simply be merged with the base model's weights as the adapter swapping would create overhead and requests using different adapters could not be batched. Rather, the LoRA computations have to be separated from the base LLM computations, and in a multi-device setup the LoRA adapters can be sharded in a way that is well aligned with the base model's tensor parallel execution, as proposed in S-LoRA. However, the S-LoRA sharding strategy encounters some communication overhead, which may be small in theory, but can be large in practice. In this paper, we propose to constrain certain LoRA factors to be block-diagonal, which allows for an alternative way of sharding LoRA adapters that does not require any additional communication for the LoRA computations. We demonstrate in extensive experiments that our block-diagonal LoRA approach is similarly parameter efficient as standard LoRA (i.e., for a similar number of parameters it achieves similar downstream performance) and that it leads to significant end-to-end speed-up over S-LoRA. For example, when serving on eight A100 GPUs, we observe up to 1.79x (1.23x) end-to-end speed-up with 0.87x (1.74x) the number of adapter parameters for Llama-3.1-70B, and up to 1.63x (1.3x) end-to-end speed-up with 0.86x (1.73x) the number of adapter parameters for Llama-3.1-8B.

cs.LG↗
arXiv2025

Dual radar-guided glide path error correction based on the Izhikevich neuron model

Aiming at the ranging and angle measurement errors caused by target reflection characteristics and system noise in dual radar tracking, this paper proposes a dual radar track error correction method based on the Izhikevich neural model. The network uses the dynamic differential equation of the Izhikevich model to simulate the discharge characteristics of biological neurons. Its input layer integrates the coordinate measurement data of the dual radar, and the output layer represents the error compensation amount through the pulse emission frequency. The spike-timing-dependent plasticity (STDP) is used to adjust the neuron connection weights dynamically, and the trajectory distortion caused by system noise and radar ranging and angle measurement errors can be effectively suppressed.

eess.SP↗
arXiv2025

AutoForge: Automated Environment Synthesis for Agentic Reinforcement Learning

Conducting reinforcement learning (RL) in simulated environments offers a cost-effective and highly scalable way to enhance language-based agents. However, previous work has been limited to semi-automated environment synthesis or tasks lacking sufficient difficulty, offering little breadth or depth. In addition, the instability of simulated users integrated into these environments, along with the heterogeneity across simulated environments, poses further challenges for agentic RL. In this work, we propose: (1) a unified pipeline for automated and scalable synthesis of simulated environments associated with high-difficulty but easily verifiable tasks; and (2) an environment level RL algorithm that not only effectively mitigates user instability but also performs advantage estimation at the environment level, thereby improving training efficiency and stability. Comprehensive evaluations on agentic benchmarks, including tau-bench, tau2-Bench, and VitaBench, validate the effectiveness of our proposed method. Further in-depth analyses underscore its out-of-domain generalization.

cs.CL↗