arXiv subjects

Weining Wang

Publications and source records attributed to Weining Wang.

At least 19 recordsLinked to original sources

arXiv2026

Primitive-Driven Compositional Forensic Visual Prompting for Open-World Face Anti-Spoofing

Open-world face anti-spoofing must address both covariate and semantic shifts: source and target domains differ in imaging conditions, while target domains contain diverse attack types absent from training. Existing prompt-based approaches often express spoofing through category semantics or language guidance, which is effective for modeling high-level concepts but is less suited to explicitly capturing the evolving fine-grained and spatially heterogeneous forensic evidence of unseen attacks. Motivated by the hypothesis that many unseen attacks can be characterized by new combinations of recurring visual cues, we propose a compositional forensic visual prompt learning framework that operates entirely in the visual feature space. Built on a frozen ViT-based vision foundation model, the framework employs patch-aware attention to refine a shared set of learnable micro-forensic primitives into localized forensic evidence units derived from image patches. Class-specific global contextual prompts then provide input-dependent routing weights that adaptively select and compose these primitives into compositional forensic visual prompts for real/spoof discrimination. The primitives are not assigned predefined semantic meanings; instead, their specialization and reuse emerge from shared parameterization and joint optimization across categories. Extensive experiments on nine open-world protocols demonstrate state-of-the-art performance, strong cross-domain generalization, and robust adaptation to unseen attacks.

cs.CV↗

arXiv2026

Experimental Observation of Ghost Image Revivals via Structured Coherence

Ghost imaging retrieves an object's image from intensity correlations between two light beams, neither of which independently carries information about the object. However, conventional ghost imaging critically relies on precise object positioning and conjugate matching between the two arms, causing the image to disappear when these conditions are violated and the object is accessible from one plane only. In this Letter, we break this fundamental limitation by reporting the first experimental observation of revivals of ghost images of arbitrary objects via engineering the longitudinal intensity autocorrelation of a structured thermal light field into a coherence comb. With the reference arm fixed, translating the object produces periodic revivals of the ghost image whenever the object position matches a comb-tooth position, reminiscent of the Talbot effect. This capability enhances the flexibility of correlation imaging, enabling robust tomographic imaging of moving and non-periodic complex objects.

physics.optics↗

arXiv2026

Estimating Network Spillovers under Dense Measurement Error

This paper analyzes spillover effects in spatial (network) models when the neighborhood (adjacency) matrix is contaminated by measurement error from reporting, aggregation, or disclosure imperfections, leading to inconsistent estimation of network effects. We introduce a regularization framework for the latent network that allows for sparse and/or low-rank structure and accommodates potential correlation between measurement errors and outcomes. We propose two estimators: (i) a two-stage procedure that first denoises the adjacency matrix and then incorporates the purified network into a regression analysis, and (ii) a Generalized Method of Moments (GMM) estimator that jointly estimates regression parameters and refines the network structure. We then establish strictly improved consistency rates for the spillover effect estimator relative to naive estimation ignoring measurement error. Simulations demonstrate that, in the presence of noisy networks, our approach reduces the root mean squared error of spillover estimates relative to conventional methods by approximately $50-80\%$. We apply our framework to examine the international spillover of economic growth, and the tax competition across U.S. states, illustrating that denoising might restore Leontief stability and yields improved estimates of spillovers.

econ.EM↗

arXiv2026

Radially correlated partially coherent beams with a deterministic vortex structure

Partially coherent beams have attracted considerable attention due to their intrinsic resilience against complex environmental perturbations. However, the intrinsic wavefront fluctuations make it fundamentally challenging to preserve well-defined orbital angular momentum during propagation. In this work, we propose and experimentally demonstrate a class of radially correlated, partially coherent beams that carry deterministic vortex structures, generated via optical conformal mapping from Cartesian to log-polar coordinates. The resulting beams exhibit a ring-shaped coherence distribution, characterized by low coherence in the radial direction and high coherence in the azimuthal direction. This unique feature of such a beam supports a well-defined deterministic vortex phase, thereby enabling the beam to preserve its ring-shaped coherence distribution during propagation through a focusing system. Our results provide new insights into the design of new partially coherent beams and may facilitate the development of applications in optical encoding, free-space information transmission, and ultrafast light-matter interactions.

physics.optics↗

arXiv2026

From Vector Autoregressions to AI-based Time Series Forecasting: A Review

Forecasting is a central goal of time-series analysis. This review centers on three major developments in recent AI-based time-series forecasting: transformers, large pretrained models for zero-shot forecasting, and diffusion-based generative forecasters. We connect these methods to the econometric tradition built around the vector autoregression (VAR) through a common object: the conditional distribution of the future given the past. The review is organized around three long-standing challenges: \emph{high dimensionality}, \emph{nonstationarity}, and \emph{nonlinearity}. We argue that modern methods make progress by expanding the classical forecasting template: they allow more flexible dynamics, use larger information sets and training corpora, and represent richer predictive distributions. Yet they often lack the inferential and structural tools that make classical models useful for testing, explanation, and policy analysis. We close by outlining open problems where econometric tools remain important.

econ.EM↗

arXiv2026

AGC: Adaptive Geodesic Correction for Adversarial Robustness on Vision-Language Models

Vision-language models like CLIP have demonstrated remarkable zero-shot transfer capabilities. However, their susceptibility to imperceptible adversarial perturbations remains a critical security concern. While test-time defenses offer a pragmatic solution for deployed models, existing approaches typically rely on gradient-based optimization during inference, incurring significant computational overhead. In this paper, we revisit the role of data augmentation in CLIP robustness and observe that augmentations are not equally effective: specific augmentations consistently provide robust geometric cues that align with correct class semantics in the hyperspherical feature space. Based on this, we propose Adaptive Geodesic Correction (AGC), a training-free defense mechanism that requires no parameter updates. AGC identifies a reliable augmentation as a geometric anchor and corrects the input feature towards it, utilizing an adaptive step size to balance robustness against clean accuracy preservation. AGC achieves superior performance across eight fine-grained datasets and three CLIP backbones, improving average robust accuracy by 44.4\% over state-of-the-art baseline while delivering a 10$\times$ reduction in inference latency. Our findings reveal a fundamental geometric property of CLIP features, offering a highly efficient and effective paradigm for robust multimodal deployment.

cs.CV↗

arXiv2026

Causal State-Dependent Local Projections

State-dependent local projections (LPs) are widely used to study how causal effects vary as a function of economic states, but shock exogeneity alone does not identify this response function. We show that identification follows when the underlying conditional mean is linear in the shock with a state-dependent coefficient, a condition satisfied in canonical micro-macro environments, including first-order perturbation solutions of heterogeneous-agent and macro-finance models. Even then, standard linear-interaction LPs generally recover only a projection of the response function, motivating LPs with nonparametric state dependence. We develop a sieve estimator and establish pointwise and uniform inference for micro-macro panels, where a distinctive challenge is that the estimator can converge at different rates across the state space. Applied to firm investment, the method uncovers a hump-shaped response to monetary policy shocks and shows that standard linear-interaction LPs substantially understate the aggregate role of financial heterogeneity.

econ.EM↗

arXiv2026

Towards High Fidelity Face Swapping: A Comprehensive Survey and New Benchmark

Face swapping has witnessed significant progress in recent years, largely driven by advances in deep generative models such as GANs and diffusion models.Despite these advances, existing methods remain fragmented across different paradigms, and their evaluation is highly inconsistent due to the lack of standardized datasets and protocols. Moreover, prior surveys primarily focus on broader deepfake generation or detection, leaving face swapping insufficiently studied as a standalone problem. In this paper, we present a comprehensive survey and benchmark for face swapping. We provide a structured review of existing methods, organizing them into five major paradigms and systematically analyzing their design principles, strengths, and limitations. To enable fair and controlled evaluation, we introduce CASIA FaceSwapping, a high-quality benchmark with balanced demographic distributions and explicit attribute variations, and establish standardized protocols to assess the robustness of different face swapping methods. Extensive experiments on representative approaches yield new insights into the performance characteristics and limitations of current techniques. Overall, our work provides a unified perspective and a principled evaluation framework to facilitate the development of more robust and controllable face swapping methods. More results can be found at https://github.com/CASIA-NLPRAI/face-swapping-survey.

cs.CV↗

arXiv2026

Agentic-MME: What Agentic Capability Really Brings to Multimodal Intelligence?

Multimodal Large Language Models (MLLMs) are evolving from passive observers into active agents, solving problems through Visual Expansion (invoking visual tools) and Knowledge Expansion (open-web search). However, existing evaluations fall short: they lack flexible tool integration, test visual and search tools separately, and evaluate primarily by final answers. Consequently, they cannot verify if tools were actually invoked, applied correctly, or used efficiently. To address this, we introduce Agentic-MME, a process-verified benchmark for Multimodal Agentic Capabilities. It contains 418 real-world tasks across 6 domains and 3 difficulty levels to evaluate capability synergy, featuring over 2,000 stepwise checkpoints that average 10+ person-hours of manual annotation per task. Each task includes a unified evaluation framework supporting sandboxed code and APIs, alongside a human reference trajectory annotated with stepwise checkpoints along dual-axis: S-axis and V-axis. To enable true process-level verification, we audit fine-grained intermediate states rather than just final answers, and quantify efficiency via an overthinking metric relative to human trajectories. Experimental results show the best model, Gemini3-pro, achieves 56.3% overall accuracy, which falls significantly to 23.0% on Level-3 tasks, underscoring the difficulty of real-world multimodal agentic problem solving.

cs.AI↗

arXiv2026

Imaging magnetically driven astrospheres: a forward modelling approach

An astrosphere is a vast, tailed bubble-like volume around a star, formed through the interaction between the stellar magnetic field, the stellar wind, and the interstellar medium (ISM). Detecting and characterizing astrospheres are essential for constraining stellar wind properties, understanding stellar evolution, and assessing the habitability of surrounding exoplanetary systems. Charge exchanges between ionized stellar wind particles and cold ISM hydrogen atoms populate the astrosphere with neutral hydrogen, which can leave observable signatures in the Lyman-$\alpha$ (Ly$\alpha$) line absorption profile. Previous studies have inferred stellar mass-loss rates by measuring Ly$\alpha$ absorption in stellar spectra caused by astrospheric neutral hydrogen. However, our knowledge of the global morphology of astrospheres remains limited and largely dependent on sometimes contradictory simulations. Here we investigate the feasibility of detecting Ly$\alpha$ emission generated by resonant scattering from \NH{} surrounding the star, enabling the construction of a two-dimensional map of the astrosphere. With a three-dimensional magnetohydrodynamic astrosphere model, we perform forward modelling of the Ly$\alpha$ emission and assess the observation feasibility according to the observational limits of the {\it Hubble Space Telescope} (HST). We further discuss the influence of varied line-of-sight orientations and averaged ISM velocity along the line-of-sight. The spatially resolved circumstellar Ly$\alpha$ emission could provide important constraints on the astrospheric configuration and stellar wind properties, such as the bow shock standing distance, the stellar wind symmetry, and the shape of the astro-tail. Our results highlight Ly$\alpha$ astrosphere detections as a promising science case for {\it HST} and future missions such as the \textit{Habitable Worlds Observatory}.}

astro-ph.SR↗

arXiv2026

RefReward-SR: LR-Conditioned Reward Modeling for Preference-Aligned Super-Resolution

Recent advances in generative super-resolution (SR) have greatly improved visual realism, yet existing evaluation and optimization frameworks remain misaligned with human perception. Full-Reference and No-Reference metrics often fail to reflect perceptual preference, either penalizing semantically plausible details due to pixel misalignment or favoring visually sharp but inconsistent artifacts. Moreover, most SR methods rely on ground-truth (GT)-dependent distribution matching, which does not necessarily correspond to human judgments. In this work, we propose RefReward-SR, a low-resolution (LR) reference-aware reward model for preference-aligned SR. Instead of relying on GT supervision or NR evaluation, RefReward-SR assesses high-resolution (HR) reconstructions conditioned on their LR inputs, treating the LR image as a semantic anchor. Leveraging the visual-linguistic priors of a Multimodal Large Language Models (MLLM), it evaluates semantic consistency and plausibility in a reasoning-aware manner. To support this paradigm, we construct RefSR-18K, the first large-scale LR-conditioned preference dataset for SR, providing pairwise rankings based on LR-HR consistency and HR naturalness. We fine-tune the MLLM with Group Relative Policy Optimization (GRPO) using LR-conditioned ranking rewards, and further integrate GRPO into SR model training with RefReward-SR as the core reward signal for preference-aligned generation. Extensive experiments show that our framework achieves substantially better alignment with human judgments, producing reconstructions that preserve semantic consistency while enhancing perceptual plausibility and visual naturalness. Code, models, and datasets will be released upon paper acceptance.

cs.CV↗

arXiv2026

Transformer-based CoVaR: Systemic Risk in Textual Information

Conditional Value-at-Risk (CoVaR) quantifies systemic financial risk by measuring the loss quantile of one asset, conditional on another asset experiencing distress. We develop a Transformer-based methodology that integrates financial news articles directly with market data to improve CoVaR estimates. Unlike approaches that use predefined sentiment scores, our method incorporates raw text embeddings generated by a large language model (LLM). We prove explicit error bounds for our Transformer CoVaR estimator, showing that accurate CoVaR learning is possible even with small datasets. Using U.S. market returns and Reuters news items from 2006--2013, our out-of-sample results show that textual information impacts the CoVaR forecasts. With better predictive performance, we identify a pronounced negative dip during market stress periods across several equity assets when comparing the Transformer-based CoVaR to both the CoVaR without text and the CoVaR using traditional sentiment measures. Our results show that textual data can be used to effectively model systemic risk without requiring prohibitively large data sets.

econ.EM↗

arXiv2025

3SGen: Unified Subject, Style, and Structure-Driven Image Generation with Adaptive Task-specific Memory

Recent image generation approaches often address subject, style, and structure-driven conditioning in isolation, leading to feature entanglement and limited task transferability. In this paper, we introduce 3SGen, a task-aware unified framework that performs all three conditioning modes within a single model. 3SGen employs an MLLM equipped with learnable semantic queries to align text-image semantics, complemented by a VAE branch that preserves fine-grained visual details. At its core, an Adaptive Task-specific Memory (ATM) module dynamically disentangles, stores, and retrieves condition-specific priors, such as identity for subjects, textures for styles, and spatial layouts for structures, via a lightweight gating mechanism along with several scalable memory items. This design mitigates inter-task interference and naturally scales to compositional inputs. In addition, we propose 3SGen-Bench, a unified image-driven generation benchmark with standardized metrics for evaluating cross-task fidelity and controllability. Extensive experiments on our proposed 3SGen-Bench and other public benchmarks demonstrate our superior performance across diverse image-driven generation tasks.

cs.CV↗

arXiv2025

TTP: Test-Time Padding for Adversarial Detection and Robust Adaptation on Vision-Language Models

Vision-Language Models (VLMs), such as CLIP, have achieved impressive zero-shot recognition performance but remain highly susceptible to adversarial perturbations, posing significant risks in safety-critical scenarios. Previous training-time defenses rely on adversarial fine-tuning, which requires labeled data and costly retraining, while existing test-time strategies fail to reliably distinguish between clean and adversarial inputs, thereby preventing both adversarial robustness and clean accuracy from reaching their optimum. To address these limitations, we propose Test-Time Padding (TTP), a lightweight defense framework that performs adversarial detection followed by targeted adaptation at inference. TTP identifies adversarial inputs via the cosine similarity shift between CLIP feature embeddings computed before and after spatial padding, yielding a universal threshold for reliable detection across architectures and datasets. For detected adversarial cases, TTP employs trainable padding to restore disrupted attention patterns, coupled with a similarity-aware ensemble strategy for a more robust final prediction. For clean inputs, TTP leaves them unchanged by default or optionally integrates existing test-time adaptation techniques for further accuracy gains. Comprehensive experiments on diverse CLIP backbones and fine-grained benchmarks show that TTP consistently surpasses state-of-the-art test-time defenses, delivering substantial improvements in adversarial robustness without compromising clean accuracy. The code for this paper will be released soon.

cs.CV↗

arXiv2025

ProAV-DiT: A Projected Latent Diffusion Transformer for Efficient Synchronized Audio-Video Generation

Sounding Video Generation (SVG) remains a challenging task due to the inherent structural misalignment between audio and video, as well as the high computational cost of multimodal data processing. In this paper, we introduce ProAV-DiT, a Projected Latent Diffusion Transformer designed for efficient and synchronized audio-video generation. To address structural inconsistencies, we preprocess raw audio into video-like representations, aligning both the temporal and spatial dimensions between audio and video. At its core, ProAV-DiT adopts a Multi-scale Dual-stream Spatio-Temporal Autoencoder (MDSA), which projects both modalities into a unified latent space using orthogonal decomposition, enabling fine-grained spatiotemporal modeling and semantic alignment. To further enhance temporal coherence and modality-specific fusion, we introduce a multi-scale attention mechanism, which consists of multi-scale temporal self-attention and group cross-modal attention. Furthermore, we stack the 2D latents from MDSA into a unified 3D latent space, which is processed by a spatio-temporal diffusion Transformer. This design efficiently models spatiotemporal dependencies, enabling the generation of high-fidelity synchronized audio-video content while reducing computational overhead. Extensive experiments conducted on standard benchmarks demonstrate that ProAV-DiT outperforms existing methods in both generation quality and computational efficiency.

cs.MM↗

arXiv2025

PhysCorr: Dual-Reward DPO for Physics-Constrained Text-to-Video Generation with Automated Preference Selection

Recent advances in text-to-video generation have achieved impressive perceptual quality, yet generated content often violates fundamental principles of physical plausibility - manifesting as implausible object dynamics, incoherent interactions, and unrealistic motion patterns. Such failures hinder the deployment of video generation models in embodied AI, robotics, and simulation-intensive domains. To bridge this gap, we propose PhysCorr, a unified framework for modeling, evaluating, and optimizing physical consistency in video generation. Specifically, we introduce PhysicsRM, the first dual-dimensional reward model that quantifies both intra-object stability and inter-object interactions. On this foundation, we develop PhyDPO, a novel direct preference optimization pipeline that leverages contrastive feedback and physics-aware reweighting to guide generation toward physically coherent outputs. Our approach is model-agnostic and scalable, enabling seamless integration into a wide range of video diffusion and transformer-based backbones. Extensive experiments across multiple benchmarks demonstrate that PhysCorr achieves significant improvements in physical realism while preserving visual fidelity and semantic alignment. This work takes a critical step toward physically grounded and trustworthy video generation.

cs.CV↗

arXiv2025

UniAlignment: Semantic Alignment for Unified Image Generation, Understanding, Manipulation and Perception

The remarkable success of diffusion models in text-to-image generation has sparked growing interest in expanding their capabilities to a variety of multi-modal tasks, including image understanding, manipulation, and perception. These tasks require advanced semantic comprehension across both visual and textual modalities, especially in scenarios involving complex semantic instructions. However, existing approaches often rely heavily on vision-language models (VLMs) or modular designs for semantic guidance, leading to fragmented architectures and computational inefficiency. To address these challenges, we propose UniAlignment, a unified multimodal generation framework within a single diffusion transformer. UniAlignment introduces a dual-stream diffusion training strategy that incorporates both intrinsic-modal semantic alignment and cross-modal semantic alignment, thereby enhancing the model's cross-modal consistency and instruction-following robustness. Additionally, we present SemGen-Bench, a new benchmark specifically designed to evaluate multimodal semantic consistency under complex textual instructions. Extensive experiments across multiple tasks and benchmarks demonstrate that UniAlignment outperforms existing baselines, underscoring the significant potential of diffusion models in unified multimodal generation.

cs.CV↗

arXiv2025

Plausible GMM: A Quasi-Bayesian Approach

Structural estimation in economics often makes use of models formulated in terms of moment conditions. While these moment conditions are generally well-motivated, it is often unknown whether the moment restrictions hold exactly. We consider a framework where researchers model their belief about the potential degree of misspecification via a prior distribution and adopt a quasi-Bayesian approach for performing inference on structural parameters. We provide quasi-posterior concentration results, verify that quasi-posteriors can be used to obtain approximately optimal Bayesian decision rules under the maintained prior structure over misspecification, and provide a form of frequentist coverage results. We illustrate the approach through empirical examples where we obtain informative inference for structural objects allowing for substantial relaxations of the requirement that moment conditions hold exactly.

econ.EM↗