arXiv subjects

Zhizhen Zhang

Publications and source records attributed to Zhizhen Zhang.

13 recordsLinked to original sources

arXiv2026

ISOMORPH: A Supply Chain Digital Twin for Simulation, Dataset Generation, and Forecasting Benchmarks

Open time-series forecasting (TSF) benchmarks cover retail, energy, weather, and traffic, but supply-chain logistics remains underserved. We introduce ISOMORPH, the first public digital twin of a multi-echelon logistics network with interpretable, user-configurable parameters and modular topology, demand, and control rules. The simulator advances a directed routing graph in discrete time: demand is served from inventory or recorded as backlog and triggers replenishment throughout the network. The state tracks inventory, outstanding orders, in-transit shipments, and a smoothed demand estimate, yielding Markovian dynamics on a tractable state space. The released data reproduces the bullwhip effect at empirically consistent magnitudes, while three conservation laws provide verification tools for simulator extensions. We release datasets at two catalogue scales ($C=50$ and $C=200$), with a 33-rollout scenario library at $C=50$. These datasets exhibit dynamics largely absent from fixed TSF benchmarks, including variance amplification, cascading bottlenecks, regime shifts, and cross-channel coupling through shared macro shocks. Zero-shot evaluation of three foundation models (Chronos, Moirai, TimesFM) against three in-domain-trained baselines (ARIMA, ETS, PatchTST) spans four targets: demand, backlog, fill rate, and edge utilization. Comparison with ETTh1, Electricity, and Weather shows that ISOMORPH introduces forecasting regimes that differ from standard real-world TSF benchmarks, positioning it as a complementary, regenerable logistics-domain benchmark. The same pairing produces forecast confidence bands across scenario configurations, providing forward UQ from parameter uncertainty and demonstrating foundation models as fast surrogates for digital-twin-based UQ. Code (MIT): https://github.com/tuhinsahai/ISOMORPH. Interactive demo: https://huggingface.co/spaces/HyeminGu/ISOMORPH-demo.

stat.ML↗

arXiv2026

Fine-Tuning Generative Models for Extreme Events via CVaR-Penalized Wasserstein Gradient Flows

We propose CVaR-penalized Generative Particle Algorithm (CVaR-GPA), a robust, tail-agnostic algorithm for fine-tuning generative models to learn heavy-tailed distributions and capture extreme events, requiring no prior knowledge or estimation of the target's tail characteristics. The method is the Wasserstein gradient flow of the Lipschitz-regularized Kullback-Leibler (KL) divergence penalized by a Conditional Value-at-Risk (CVaR) discrepancy term: the Lipschitz-regularized KL divergence enables robust learning under minimal assumptions on the target distribution, while the CVaR penalty restores the velocity that otherwise vanishes prematurely in the under-sampled tails. The penalized flow admits a bounded but non-Lipschitz velocity field. This departs from the Lipschitz transport maps of standard generators, which preserve the tail behavior of a light-tailed source, and enables transport toward heavier-tailed targets. To define this flow on empirical measures, we derive the first-variation subgradients of CVaR from its Rockafellar-Uryasev representation, valid precisely where the classical density-based formula fails. The particle algorithm CVaR-GPA fine-tunes the output samples of any pre-trained model, without access to its architecture, and runs on an adaptive time horizon set by a kinetic-energy stopping criterion rather than a preset depth. On synthetic isotropic and anisotropic Student-$t$ target distributions, Neal's funnel distribution, and the real-world high-dimensional Fama-French 25 portfolio dataset, CVaR-GPA dramatically improves global and tail accuracy on heavy-tailed targets over the pre-trained baseline.

stat.ML↗

arXiv2026

StableI2I: Spotting Unintended Changes in Image-to-Image Transition

In most real-world image-to-image (I2I) scenarios, existing evaluations primarily focus on instruction following and the perceptual quality or aesthetics of the generated images. However, they largely fail to assess whether the output image preserves the semantic correspondence and spatial structure of the input image. To address this limitation, we propose StableI2I, a unified and dynamic evaluation framework that explicitly measures content fidelity and pre--post consistency across a wide range of I2I tasks without requiring reference images, including image editing and image restoration. In addition, we construct StableI2I-Bench, a benchmark designed to systematically evaluate the accuracy of MLLMs on such fidelity and consistency assessment tasks. Extensive experimental results demonstrate that StableI2I provides accurate, fine-grained, and interpretable evaluations of content fidelity and consistency, with strong correlations to human subjective judgments. Our framework serves as a practical and reliable evaluation tool for diagnosing content consistency and benchmarking model performance in real-world I2I systems.

cs.CV↗

arXiv2026

AnchorVLA: Anchored Diffusion for Efficient End-to-End Mobile Manipulation

A central challenge in mobile manipulation is preserving multiple plausible action models while remaining reactive during execution. A bottle in a cluttered scene can often be approached and grasped in multiple valid ways. Robust behavior depends on preserving this action diversity while remaining reactive as the scene evolves. Diffusion policies are appealing because they model multimodal action distributions rather than collapsing to one solution. But in practice, full iterative denoising is costly at control time. Action chunking helps amortize inference, yet it also creates partially open-loop behavior, allowing small mismatches to accumulate into drift. We present AnchorVLA, a diffusion-based VLA policy for mobile manipulation built on the core insight that when sampling begins near a plausible solution manifold, extensive denoising is unnecessary to recover multimodal, valid actions. AnchorVLA combines a lightweight VLA adaptation backbone with an anchored diffusion action head, which denoises locally around anchor trajectories using a truncated diffusion schedule. This retains multimodal action generation while reducing inference cost for closed-loop control. Crucially, to mitigate chunking-induced drift, we introduce a test-time self-correction mechanism via a lightweight residual correction module that makes high-frequency, per-step adjustments during rollout. Across diverse mobile manipulation tasks, AnchorVLA improves success and stability under disturbances and distribution shifts while maintaining low-latency inference. The source code is made available at https://github.com/jason-lim26/AnchorVLA.

cs.RO↗

arXiv2026

MergeVLA: Cross-Skill Model Merging Toward a Generalist Vision-Language-Action Agent

Recent Vision-Language-Action (VLA) models reformulate vision-language models by tuning them with millions of robotic demonstrations. While they perform well when fine-tuned for a single embodiment or task family, extending them to multi-skill settings remains challenging: directly merging VLA experts trained on different tasks results in near-zero success rates. This raises a fundamental question: what prevents VLAs from mastering multiple skills within one model? With an empirical decomposition of learnable parameters during VLA fine-tuning, we identify two key sources of non-mergeability: (1) Finetuning drives LoRA adapters in the VLM backbone toward divergent, task-specific directions beyond the capacity of existing merging methods to unify. (2) Action experts develop inter-block dependencies through self-attention feedback, causing task information to spread across layers and preventing modular recombination. To address these challenges, we present MergeVLA, a merging-oriented VLA architecture that preserves mergeability by design. MergeVLA introduces sparsely activated LoRA adapters via task masks to retain consistent parameters and reduce irreconcilable conflicts in the VLM. Its action expert replaces self-attention with cross-attention-only blocks to keep specialization localized and composable. When the task is unknown, it uses a test-time task router to adaptively select the appropriate task mask and expert head from the initial observation, enabling unsupervised task inference. Across LIBERO, LIBERO-Plus, RoboTwin, and multi-task experiments on the real SO101 robotic arm, MergeVLA achieves performance comparable to or even exceeding individually finetuned experts, demonstrating robust generalization across tasks, embodiments, and environments. Project page: https://mergevla.github.io/

cs.RO↗

arXiv2025

Provable Ordering and Continuity in Vision-Language Pretraining for Generalizable Embodied Agents

Pre-training vision-language representations on human action videos has emerged as a promising approach to reduce reliance on large-scale expert demonstrations for training embodied agents. However, prior methods often employ time contrastive learning based on goal-reaching heuristics, progressively aligning language instructions from the initial to the final frame. This overemphasis on future frames can result in erroneous vision-language associations, as actions may terminate early or include irrelevant moments in the end. To address this issue, we propose Action Temporal Coherence Learning (AcTOL) to learn ordered and continuous vision-language representations without rigid goal-based constraint. AcTOL treats a video as a continuous trajectory where it (1) contrasts semantic differences between frames to reflect their natural ordering, and (2) imposes a local Brownian bridge constraint to ensure smooth transitions across intermediate frames. Extensive imitation learning experiments on both simulated and real robots show that the pretrained features significantly enhance downstream manipulation tasks with high robustness to different linguistic styles of instructions, offering a viable pathway toward generalized embodied agents.

cs.RO↗

arXiv2024

Contrastive Learning for Implicit Social Factors in Social Media Popularity Prediction

On social media sharing platforms, some posts are inherently destined for popularity. Therefore, understanding the reasons behind this phenomenon and predicting popularity before post publication holds significant practical value. The previous work predominantly focuses on enhancing post content extraction for better prediction results. However, certain factors introduced by social platforms also impact post popularity, which has not been extensively studied. For instance, users are more likely to engage with posts from individuals they follow, potentially influencing the popularity of these posts. We term these factors, unrelated to the explicit attractiveness of content, as implicit social factors. Through the analysis of users' post browsing behavior (also validated in public datasets), we propose three implicit social factors related to popularity, including content relevance, user influence similarity, and user identity. To model the proposed social factors, we introduce three supervised contrastive learning tasks. For different task objectives and data types, we assign them to different encoders and control their gradient flows to achieve joint optimization. We also design corresponding sampling and augmentation algorithms to improve the effectiveness of contrastive learning. Extensive experiments on the Social Media Popularity Dataset validate the superiority of our proposed method and also confirm the important role of implicit social factors in popularity prediction. We open source the code at https://github.com/Daisy-zzz/PPCL.git.

cs.SI↗

arXiv2024

Similarity Guided Multimodal Fusion Transformer for Semantic Location Prediction in Social Media

Semantic location prediction aims to derive meaningful location insights from multimodal social media posts, offering a more contextual understanding of daily activities than using GPS coordinates. This task faces significant challenges due to the noise and modality heterogeneity in "text-image" posts. Existing methods are generally constrained by inadequate feature representations and modal interaction, struggling to effectively reduce noise and modality heterogeneity. To address these challenges, we propose a Similarity-Guided Multimodal Fusion Transformer (SG-MFT) for predicting the semantic locations of users from their multimodal posts. First, we incorporate high-quality text and image representations by utilizing a pre-trained large vision-language model. Then, we devise a Similarity-Guided Interaction Module (SIM) to alleviate modality heterogeneity and noise interference by incorporating both coarse-grained and fine-grained similarity guidance for improving modality interactions. Specifically, we propose a novel similarity-aware feature interpolation attention mechanism at the coarse-grained level, leveraging modality-wise similarity to mitigate heterogeneity and reduce noise within each modality. At the fine-grained level, we utilize a similarity-aware feed-forward block and element-wise similarity to further address the issue of modality heterogeneity. Finally, building upon pre-processed features with minimal noise and modal interference, we devise a Similarity-aware Fusion Module (SFM) to fuse two modalities with a cross-attention mechanism. Comprehensive experimental results clearly demonstrate the superior performance of our proposed method.

cs.CV↗

arXiv2024

HierCas: Hierarchical Temporal Graph Attention Networks for Popularity Prediction in Information Cascades

Information cascade popularity prediction is critical for many applications, including but not limited to identifying fake news and accurate recommendations. Traditional feature-based methods heavily rely on handcrafted features, which are domain-specific and lack generalizability to new domains. To address this problem, researchers have turned to neural network-based approaches. However, most existing methods follow a sampling-based modeling approach, potentially losing continuous dynamic information that emerges during the information diffusion process. In this paper, we propose Hierarchical Temporal Graph Attention Networks for cascade popularity prediction (HierCas), which operates on the entire cascade graph by a dynamic graph modeling approach. By leveraging time-aware node embedding, graph attention mechanisms, and hierarchical pooling structures, HierCas effectively captures the popularity trend implicit in the complex cascade. Extensive experiments conducted on two real-world datasets in different scenarios demonstrate that our HierCas significantly outperforms the state-of-the-art approaches. We have released our code at https://github.com/Daisy-zzz/HierCas.

cs.SI↗

arXiv2024

View while Moving: Efficient Video Recognition in Long-untrimmed Videos

Recent adaptive methods for efficient video recognition mostly follow the two-stage paradigm of "preview-then-recognition" and have achieved great success on multiple video benchmarks. However, this two-stage paradigm involves two visits of raw frames from coarse-grained to fine-grained during inference (cannot be parallelized), and the captured spatiotemporal features cannot be reused in the second stage (due to varying granularity), being not friendly to efficiency and computation optimization. To this end, inspired by human cognition, we propose a novel recognition paradigm of "View while Moving" for efficient long-untrimmed video recognition. In contrast to the two-stage paradigm, our paradigm only needs to access the raw frame once. The two phases of coarse-grained sampling and fine-grained recognition are combined into unified spatiotemporal modeling, showing great performance. Moreover, we investigate the properties of semantic units in video and propose a hierarchical mechanism to efficiently capture and reason about the unit-level and video-level temporal semantics in long-untrimmed videos respectively. Extensive experiments on both long-untrimmed and short-trimmed videos demonstrate that our approach outperforms state-of-the-art methods in terms of accuracy as well as efficiency, yielding new efficiency and accuracy trade-offs for video spatiotemporal modeling.

cs.CV↗

arXiv2023

Traffic flow phase transition phenomena based on the kinetic approach

We develop a discrete Boltzmann-type model that uses dynamics in phase space to describe the behavior of traffic flows. Firstly, we model the traffic flow at mesoscopic scale using dynamics in phase space, which is considered as an evolution model in a graph, and we demonstrate the existence of phase transition phenomena through theoretical analysis. Secondly, we assumed the density of vehicles in geometric space to be homogeneous and single-peaked, respectively, and performed numerical simulations to obtain the results consistent with experience. According to this model, we can perform effective simulations on the effects played by multiple parameters on the traffic flow.

nlin.CG↗

arXiv2023

Improving Social Media Popularity Prediction with Multiple Post Dependencies

Social Media Popularity Prediction has drawn a lot of attention because of its profound impact on many different applications, such as recommendation systems and multimedia advertising. Despite recent efforts to leverage the content of social media posts to improve prediction accuracy, many existing models fail to fully exploit the multiple dependencies between posts, which are important to comprehensively extract content information from posts. To tackle this problem, we propose a novel prediction framework named Dependency-aware Sequence Network (DSN) that exploits both intra- and inter-post dependencies. For intra-post dependency, DSN adopts a multimodal feature extractor with an efficient fine-tuning strategy to obtain task-specific representations from images and textual information of posts. For inter-post dependency, DSN uses a hierarchical information propagation method to learn category representations that could better describe the difference between posts. DSN also exploits recurrent networks with a series of gating layers for more flexible local temporal processing abilities and multi-head attention for long-term dependencies. The experimental results on the Social Media Popularity Dataset demonstrate the superiority of our method compared to existing state-of-the-art models.

cs.MM↗

arXiv2021

The dynamic energy balance in earthquakes expressed by fault surface morphology

The dynamic energy balance is essential for earthquake studies. The energy balance approach is one of the most famous developments in fracture mechanics. To interpret seismological data, crack models and sliding on a frictional surface (fault) models are widely used. The macroscopically observable energy budget and the microscopic processes can be related through the fracture energy $G_c$. The fault surface morphology is the direct result of the microscopic processes near the crack tip or on the frictional interface. Here we show that the dynamic energy balance in earthquakes can be expressed by fault surface morphology, and that they are quantitatively linked. The direct shear experiments proves the predictions of the theoretical discussions, and show that the strain rate has crucial influence on the dynamic energy balance.

physics.geo-ph↗