arXiv subjects

Wenjun Mei

Publications and source records attributed to Wenjun Mei.

At least 19 recordsLinked to original sources

arXiv2026

A Minimal Dynamical Model for Incubation-Outbreak Transitions in Social Norm Diffusion

In this paper, we introduce a minimal dynamical model for the diffusion of a new social norm, in which individuals transition among three states: non-supporters, silent supporters, and vocal advocates. Despite its simplicity and its close relation to the classical SI-type and SIS-type spreading dynamics, this model exhibits a nontrivial latent-outbreak dynamic pattern: an initial small adoption wave is followed by a long quiescent period and then an abrupt, endogenous explosion of support. Through a complete analytical characterization of equilibria, stability, convergence, and phase-transition conditions, we identify the nonlinear mechanism that generates this incubation phenomenon and derive estimates of the latent period. Our results reveal a dynamical route to sudden social change driven purely by internal interactions rather than external shocks.

eess.SY↗

arXiv2026

Forgetting-Factor Regret for Online Zero-Sum Games

This paper studies dynamic equilibrium tracking in online two-player zero-sum games with time-varying convex-concave payoff functions. Existing regret metrics for online saddle-point problems usually aggregate historical payoffs with uniform weights, and hence may fail to characterize the real-time tracking performance with respect to the current Nash equilibrium (NE). To address this issue, we introduce a zero-sum game regret function with a forgetting factor, which assigns exponentially decaying weights to past saddle gaps and emphasizes recent performance. This metric directly links regret minimization to the tracking of time-varying NEs. Within this framework, we investigate three online algorithms under different computational and information settings. For first-order feedback, we analyze projected gradient descent-ascent and design a projection-free online Frank-Wolfe method to reduce the computational cost of projections. For zeroth-order feedback, we develop a deterministic finite-difference method that only uses function-value queries. For all three algorithms, we establish forgetting-factor regret bounds that explicitly characterize the effects of NE variation, payoff variation, and gradient-estimation error. We further provide sufficient conditions under which the proposed regret converges to zero, thereby certifying asymptotic tracking of time-varying NEs. The numerical example validates the theoretical results and illustrates the tracking advantage of the proposed regret metric.

math.OC↗

arXiv2026

GigaWorld-Policy: An Efficient Action-Centered World--Action Model

World-Action Models (WAM) initialized from pre-trained video generation backbones have demonstrated remarkable potential for robot policy learning. However, existing approaches face two critical bottlenecks that hinder performance and deployment. First, jointly reasoning over future visual dynamics and corresponding actions incurs substantial inference overhead. Second, joint modeling often entangles visual and motion representations, making motion prediction accuracy heavily dependent on the quality of future video forecasts. To address these issues, we introduce GigaWorld-Policy, an action-centered WAM that learns 2D pixel-action dynamics while enabling efficient action decoding, with optional video generation. Specifically, we formulate policy training into two coupled components: the model predicts future action sequences conditioned on the current observation, and simultaneously generates future videos conditioned on the predicted actions and the same observation. The policy is supervised by both action prediction and video generation, providing richer learning signals and encouraging physically plausible actions through visual-dynamics constraints. With a causal design that prevents future-video tokens from influencing action tokens, explicit future-video generation is optional at inference time, allowing faster action prediction during deployment. To support this paradigm, we curate a diverse, large-scale robot dataset to pre-train an action-centered video generation model, which is then adapted as the backbone for robot policy learning. Experimental results on real-world robotic platforms show that GigaWorld-Policy runs 9x faster than the leading WAM baseline, Motus, while improving task success rates by 7%. Moreover, compared with pi-0.5, GigaWorld-Policy improves performance by 95% on RoboTwin 2.0.

cs.CV↗

arXiv2025

SwiftVLA: Unlocking Spatiotemporal Dynamics for Lightweight VLA Models at Minimal Overhead

Vision-Language-Action (VLA) models built on pretrained Vision-Language Models (VLMs) show strong potential but are limited in practicality due to their large parameter counts. To mitigate this issue, using a lightweight VLM has been explored, but it compromises spatiotemporal reasoning. Although some methods suggest that incorporating additional 3D inputs can help, they usually rely on large VLMs to fuse 3D and 2D inputs and still lack temporal understanding. Therefore, we propose SwiftVLA, an architecture that enhances a compact model with 4D understanding while preserving design efficiency. Specifically, our approach features a pretrained 4D visual geometry transformer with a temporal cache that extracts 4D features from 2D images. Then, to enhance the VLM's ability to exploit both 2D images and 4D features, we introduce Fusion Tokens, a set of learnable tokens trained with a future prediction objective to generate unified representations for action generation. Finally, we introduce a mask-and-reconstruct strategy that masks 4D inputs to the VLM and trains the VLA to reconstruct them, enabling the VLM to learn effective 4D representations and allowing the 4D branch to be dropped at inference with minimal performance loss. Experiments in real and simulated environments show that SwiftVLA outperforms lightweight baselines and rivals VLAs up to 7 times larger, achieving comparable performance on edge devices while being 18 times faster and reducing memory footprint by 12 times.

cs.CV↗

arXiv2025

Power Allocation Games on Signed Networks: Nash Equilibria and Coevolutionary Dynamics

Understanding how strategic interactions and power distributions coevolve in international relations is central to explaining conflict, cooperation, and long-term inequality. We study this problem using a power-allocation game on signed networks. Departing from models that restrict strategy updates to Pareto improvements, we propose a generalized formulation in which countries prioritize self-survival and strategically trade off between supporting allies and weakening adversaries. This relaxation allows countries to sacrifice certain allies to achieve higher overall payoffs. For the resulting static game, we establish the existence of pure-strategy Nash equilibria and characterize their properties in extreme cases, including fully antagonistic networks and the presence of a dominant power. We further introduce a power-strategy coevolutionary dynamic and prove its almost-sure convergence to equilibria corresponding to the static game. The proposed models are validated using empirical data and numerical simulations. Historical data from the Correlates of War and national capability datasets show that survival likelihood predicts countries' safety outcomes and subsequent economic growth with relatively high accuracy. Simulations further indicate that, under fixed conflict intensity, more structurally balanced signed networks yield higher average power and lower inequality at steady states.

cs.GT↗

arXiv2025

Pareto-Improvement-Driven Opinion Dynamics Explaining the Emergence of Pluralistic Ignorance

Opinion dynamics has recently been modeled from a game-theoretic perspective, where opinion updates are captured by individuals' cost functions representing their motivations. Conventional formulations aggregate multiple motivations into a single objective, implicitly assuming that these motivations are interchangeable. This paper challenges that assumption and proposes an opinion dynamics model grounded in a multi-objective game framework. In the proposed model, each individual experiences two distinct costs: social pressure from disagreement with others and cognitive dissonance from deviation from the perceived truth. Opinion updates are modeled as Pareto improvements between these two costs. This framework provides a parsimonious explanation for the emergence of pluralistic ignorance, where individuals may agree on something untrue even though they all know the underlying truth. We analytically characterize the model, derive conditions for the emergence and prevalence of the truth, and propose an initial-seeding strategy that ensures consensus on truth. Numerical simulations are conducted on how network density and clustering affect the expression of truth. Both theoretical and numerical results lead to clear and non-trivial sociological insights. For example, no network structure guarantees almost-sure consensus on truth if no one initially expresses the truth; moderately sparse but well-mixed networks are most conducive to consensus on truth.

eess.SY↗

arXiv2025

A Zeroth-order Resilient Algorithm for Distributed Online Optimization against Byzantine Edge Attacks

In this paper, we propose a zeroth-order resilient distributed online algorithm for networks under Byzantine edge attacks. We assume that both the edges attacked by Byzantine adversaries and the objective function are time-varying. Moreover, we focus on the scenario where the complete time-varying objective function cannot be observed, and only its value at a certain point is available. Using deterministic difference, we design a zeroth-order distributed online optimization algorithm against Byzantine edge attacks and provide an upper bound on the dynamic regret of the algorithm. Finally, a simulation example is given justifying the theoretical results.

math.OC↗

arXiv2025

Meritocracy versus Matthew-effect: Two underlying network formation mechanisms of online social platforms

With the rapid development of the internet industry, online social networks have come to play an increasingly significant role in everyday life. In recent years, content-based emerging platforms such as TikTok, Instagram, and Bilibili have diverged fundamentally in their underlying logic from traditional connection-based social platforms like Facebook and LinkedIn. Empirical data on follower counts and follower-count-based rankings reveal that the distribution of social power varies significantly across different types of platforms, with content-based platforms exhibiting notably greater inequality. Here we propose two fundamental network formation mechanisms: a meritocracy-based model and a Matthew-effect-based model, designed to capture the formation logic underlying traditional and emerging social networks, respectively. Through theoretical and numerical analysis, we demonstrate that both models replicate salient statistical features of social networks including scale-free and small-world property, while also closely match empirical patterns on the relationship between in-degrees and in-degree rankings, thereby capturing the distinctive distributions of social power in respective platforms. Moreover, networks such as academic collaboration networks, where the distribution of social power usually lies between that of traditional and emerging platorms, can be interpreted through a hybrid of the two proposed mechanisms. Deconstructing the formation mechanisms of online social networks offers valuable insights into the evolution of the content ecosystems and the behavioral patterns of content creators on online social platforms.

cs.SI↗

arXiv2025

ReconDreamer-RL: Enhancing Reinforcement Learning via Diffusion-based Scene Reconstruction

Reinforcement learning for training end-to-end autonomous driving models in closed-loop simulations is gaining growing attention. However, most simulation environments differ significantly from real-world conditions, creating a substantial simulation-to-reality (sim2real) gap. To bridge this gap, some approaches utilize scene reconstruction techniques to create photorealistic environments as a simulator. While this improves realistic sensor simulation, these methods are inherently constrained by the distribution of the training data, making it difficult to render high-quality sensor data for novel trajectories or corner case scenarios. Therefore, we propose ReconDreamer-RL, a framework designed to integrate video diffusion priors into scene reconstruction to aid reinforcement learning, thereby enhancing end-to-end autonomous driving training. Specifically, in ReconDreamer-RL, we introduce ReconSimulator, which combines the video diffusion prior for appearance modeling and incorporates a kinematic model for physical modeling, thereby reconstructing driving scenarios from real-world data. This narrows the sim2real gap for closed-loop evaluation and reinforcement learning. To cover more corner-case scenarios, we introduce the Dynamic Adversary Agent (DAA), which adjusts the trajectories of surrounding vehicles relative to the ego vehicle, autonomously generating corner-case traffic scenarios (e.g., cut-in). Finally, the Cousin Trajectory Generator (CTG) is proposed to address the issue of training data distribution, which is often biased toward simple straight-line movements. Experiments show that ReconDreamer-RL improves end-to-end autonomous driving training, outperforming imitation learning methods with a 5x reduction in the Collision Ratio.

cs.CV↗

arXiv2025

Human Mobility in Epidemic Modeling

Human mobility forms the backbone of contact patterns through which infectious diseases propagate, fundamentally shaping the spatio-temporal dynamics of epidemics and pandemics. While traditional models are often based on the assumption that all individuals have the same probability of infecting every other individual in the population, a so-called random homogeneous mixing, they struggle to capture the complex and heterogeneous nature of real-world human interactions. Recent advancements in data-driven methodologies and computational capabilities have unlocked the potential of integrating high-resolution human mobility data into epidemic modeling, significantly improving the accuracy, timeliness, and applicability of epidemic risk assessment, contact tracing, and intervention strategies. This review provides a comprehensive synthesis of the current landscape in human mobility-informed epidemic modeling. We explore diverse sources and representations of human mobility data, and then examine the behavioral and structural roles of mobility and contact in shaping disease transmission dynamics. Furthermore, the review spans a wide range of epidemic modeling approaches, ranging from classical compartmental models to network-based, agent-based, and machine learning models. And we also discuss how mobility integration enhances risk management and response strategies during epidemics. By synthesizing these insights, the review can serve as a foundational resource for researchers and practitioners, bridging the gap between epidemiological theory and the dynamic complexities of human interaction while charting clear directions for future research.

cs.SI↗

arXiv2025

Linear-Quadratic Discrete-Time Dynamic Games with Unknown Dynamics

Considering linear-quadratic discrete-time games with unknown input/output/state (i/o/s) dynamics and state, we provide necessary and sufficient conditions for the existence and uniqueness of feedback Nash equilibria (FNE) in the finite-horizon game, based entirely on offline input/output data. We prove that the finite-horizon unknown-dynamics game and its corresponding known-dynamics game have the same FNEs, and provide detailed relationships between their respective FNE matrices. To simplify the computation of FNEs, we provide an invertibility condition and a corresponding algorithm that computes one FNE by solving a finite number of linear equation systems using offline data. For the infinite-horizon unknown-dynamics game, limited offline data restricts players to computing optimal strategies only over a finite horizon. We prove that the finite-horizon strategy ``watching $T$ steps into the future and moving one step now,'' which is commonly used in classical optimal control, exhibits convergence in both the FNE matrices and the total costs in the infinite-horizon unknown-dynamics game, and further provide an analysis of the convergence rate of the total cost. The corresponding algorithm for the infinite-horizon game is proposed and its efficacy is demonstrated through a non-scalar numerical example.

eess.SY↗

arXiv2025

WonderFree: Enhancing Novel View Quality and Cross-View Consistency for 3D Scene Exploration

Interactive 3D scene generation from a single image has gained significant attention due to its potential to create immersive virtual worlds. However, a key challenge in current 3D generation methods is the limited explorability, which cannot render high-quality images during larger maneuvers beyond the original viewpoint, particularly when attempting to move forward into unseen areas. To address this challenge, we propose WonderFree, the first model that enables users to interactively generate 3D worlds with the freedom to explore from arbitrary angles and directions. Specifically, we decouple this challenge into two key subproblems: novel view quality, which addresses visual artifacts and floating issues in novel views, and cross-view consistency, which ensures spatial consistency across different viewpoints. To enhance rendering quality in novel views, we introduce WorldRestorer, a data-driven video restoration model designed to eliminate floaters and artifacts. In addition, a data collection pipeline is presented to automatically gather training data for WorldRestorer, ensuring it can handle scenes with varying styles needed for 3D scene generation. Furthermore, to improve cross-view consistency, we propose ConsistView, a multi-view joint restoration mechanism that simultaneously restores multiple perspectives while maintaining spatiotemporal coherence. Experimental results demonstrate that WonderFree not only enhances rendering quality across diverse viewpoints but also significantly improves global coherence and consistency. These improvements are confirmed by CLIP-based metrics and a user study showing a 77.20% preference for WonderFree over WonderWorld enabling a seamless and immersive 3D exploration experience. The code, model, and data will be publicly available.

cs.CV↗

arXiv2025

On finite-horizon approximation of an infinite-horizon feedback Nash equilibrium in discrete-time LQ games

In infinite-horizon discrete-time linear-quadratic (LQ) dynamic games, computing feedback Nash equilibria (FNEs) remains computationally challenging. Motivated by this, we study a finite-horizon strategy for approximating one of the infinite-horizon FNEs. The finite-horizon strategy is as follows. Each player $i$ has an individual prediction horizon $T^i$. In the infinite-horizon game, at each stage, each player $i$ computes its control in the following way: player $i$ envisions an auxiliary $T^i$-stage game in which the same set of players play, computes the unique FNE of the auxiliary game using a standard method, and implements only the first-stage control. Our main result is, under suitable conditions, the total cost under these finite-horizon strategies converges to that under one of the infinite-horizon FNEs when all players' prediction horizons tend to infinity. Moreover, we derive an explicit cubic-polynomial upper bound on this cost gap with respect to the distance between the corresponding strategy matrices. This strategy is tractable and implementable, as it avoids the direct solution of the coupled algebraic Riccati equations (CARE) of infinite-horizon LQ games.

eess.SY↗

arXiv2025

Discovering Mechanistic Causality from Time Series: A Behavioral-System Approach

Identifying ``true causality'' is a fundamental challenge in complex systems research. Widely adopted methods, like the Granger causality test, capture statistical dependencies between variables rather than genuine driver-response mechanisms. This critical gap stems from the absence of mathematical tools that reliably reconstruct underlying system dynamics from observational time-series data. In this paper, we introduce a new control-based method for causality discovery through the behavior-system theory, which represents dynamical systems via trajectory spaces and has been widely used in data-driven control. Our core contribution is the \textbf{B}ehavior-\textbf{e}nabled \textbf{Caus}ality test (the BeCaus test), which transforms causality discovery into solving fictitious control problems. By exploiting the intrinsic asymmetry between system inputs and outputs, the proposed method operationalizes our conceptualization of mechanistic causality: variable $X$ is a cause of $Y$ if $X$ (partially) drives the evolution of $Y$. We establish conditions for linear time-invariant systems to be causality-discoverable, i.e., conditions for the BeCaus test to distinguish four basic causal structures (independence, full causality, partial causality, and latent-common-cause relation). Notably, our approach accommodates open systems with unobserved inputs. Moreover, an exploratory case study indicates the new method's potential extensibility to nonlinear systems.

math.OC↗

arXiv2025

Convergence and consensus analysis of a class of best-response opinion dynamics

Opinion dynamics aims to understand how individuals' opinions evolve through local interactions. Recently, opinion dynamics have been modeled as network games, where individuals update their opinions in order to minimize the social pressure caused by disagreeing with others. In this paper, we study a class of best response opinion dynamics introduced by Mei et al., where a parameter $\alpha > 0$ controls the marginal cost of opinion differences, bridging well-known mechanisms such as the DeGroot model ($\alpha = 2$) and the weighted-median model ($\alpha = 1$). We conduct theoretical analysis on how different values of $\alpha$ affect the system's convergence and consensus behavior. For the case when $\alpha > 1$, corresponding to increasing marginal costs, we establish the convergence of the dynamics and derive graph-theoretic conditions for consensus formation, which is proved to be similar to those in the DeGroot model. When $\alpha < 1$, we show via a counterexample that convergence is not always guaranteed, and we provide sufficient conditions for convergence and consensus. Additionally, numerical simulations on small-world networks reveal how network structure and $\alpha$ together affect opinion diversity.

math.DS↗

arXiv2025

WonderTurbo: Generating Interactive 3D World in 0.72 Seconds

Interactive 3D generation is gaining momentum and capturing extensive attention for its potential to create immersive virtual experiences. However, a critical challenge in current 3D generation technologies lies in achieving real-time interactivity. To address this issue, we introduce WonderTurbo, the first real-time interactive 3D scene generation framework capable of generating novel perspectives of 3D scenes within 0.72 seconds. Specifically, WonderTurbo accelerates both geometric and appearance modeling in 3D scene generation. In terms of geometry, we propose StepSplat, an innovative method that constructs efficient 3D geometric representations through dynamic updates, each taking only 0.26 seconds. Additionally, we design QuickDepth, a lightweight depth completion module that provides consistent depth input for StepSplat, further enhancing geometric accuracy. For appearance modeling, we develop FastPaint, a 2-steps diffusion model tailored for instant inpainting, which focuses on maintaining spatial appearance consistency. Experimental results demonstrate that WonderTurbo achieves a remarkable 15X speedup compared to baseline methods, while preserving excellent spatial consistency and delivering high-quality output.

cs.CV↗

arXiv2025

Mechanism Design in Max-Flows

This paper studies allocation mechanisms in max-flow games with players' capacities as private information. We first show that no core-selection mechanism is truthful: there may exist a player whose payoff increases if she under-reports her capacity when a core-section mechanism is adopted. We then introduce five desirable properties for mechanisms in max-flow games: DSIC (truthful reporting is a dominant strategy), SIR (individual rationality and positive payoff for each player contributing positively to at least one coalition), SP (no edge has an incentive to split into parallel edges), MP (no parallel edges have incentives to merge), and CM (a player's payoff does not decrease as another player's capacity and max-flow increase). While the Shapley value mechanism satisfies DSIC and SIR, it fails to meet SP, MP and CM. We propose a new mechanism based on minimal cuts that satisfies all five properties.

cs.GT↗

arXiv2024

ReconDreamer: Crafting World Models for Driving Scene Reconstruction via Online Restoration

Closed-loop simulation is crucial for end-to-end autonomous driving. Existing sensor simulation methods (e.g., NeRF and 3DGS) reconstruct driving scenes based on conditions that closely mirror training data distributions. However, these methods struggle with rendering novel trajectories, such as lane changes. Recent works have demonstrated that integrating world model knowledge alleviates these issues. Despite their efficiency, these approaches still encounter difficulties in the accurate representation of more complex maneuvers, with multi-lane shifts being a notable example. Therefore, we introduce ReconDreamer, which enhances driving scene reconstruction through incremental integration of world model knowledge. Specifically, DriveRestorer is proposed to mitigate artifacts via online restoration. This is complemented by a progressive data update strategy designed to ensure high-quality rendering for more complex maneuvers. To the best of our knowledge, ReconDreamer is the first method to effectively render in large maneuvers. Experimental results demonstrate that ReconDreamer outperforms Street Gaussians in the NTA-IoU, NTL-IoU, and FID, with relative improvements by 24.87%, 6.72%, and 29.97%. Furthermore, ReconDreamer surpasses DriveDreamer4D with PVG during large maneuver rendering, as verified by a relative improvement of 195.87% in the NTA-IoU metric and a comprehensive user study.

cs.CV↗