arXiv subjects

Zhipeng Huang

Publications and source records attributed to Zhipeng Huang.

At least 19 recordsLinked to original sources

arXiv2026

WeLM-Image: A Unified Image Generation Model towards Visual Intelligence

We present \textbf{WeLM-Image}, a unified image foundation model that integrates multimodal understanding into text-to-image generation, instruction-guided editing, and multi-reference synthesis. Designed for single-GPU deployment in WeChat's large-scale consumer applications, the model processes text, reference images, and target latents as one interleaved sequence. A Mixture-of-Transformers (MoT) backbone combines shared attention with independent understanding and generation parameters to limit optimization interference. Equipping each path with Mixture-of-Experts (MoE) modules provides sparse capacity, with 31.7B stored parameters and approximately 5.4B generation parameters activated per target latent token. A generation-oriented data engine combines a knowledge-grounded visual vocabulary with large-scale visual clustering to balance semantic coverage and editing tasks, while training-time evaluation guides iterative data refinement. Bilingual annotations and verified editing supervision support progressive training up to native 1.5K resolution. We optimize generation and editing separately with rubric- and checklist-based rewards, then consolidate the specialized policies through noise-aware on-policy distillation. In the reported overall comparisons, the 1536-resolution model ranks first among the evaluated open-source models on PRISM-Bench, Qwen-Image-Bench, and UnicBench. MeanFlow-based few-step distillation yields WeLM-Image-Turbo, which generates a $1024\times1024$ image in 4.98\,s on a single NVIDIA H20 GPU using eight guidance-free steps.

cs.CV↗

arXiv2026

From Evaluation to Evolution: Benchmarking Image Editing and Evolving Image Models

Real-world image editing requires coordinating multiple visual transformations with content preservation and reference constraints, yet aggregate scores can readily obscure unmet local requirements. Effective evaluation must therefore identify local failures and translate them into priorities for subsequent model improvement. We introduce WeGenEditBench, comprising 2,740 evaluation instances and 148 task types across single-image, text-intensive, and multi-reference editing. Pre-generated checklists shared across models decompose each request into atomic requirements; item-level deductions and visual evidence are established first and then summarized into scores for edit execution, preservation, visual quality, and task-specific correctness. To support the experience-acquisition component of recursive self-improvement (RSI), we develop WeGenEvoStudio, a unified multi-agent harness for RSI-oriented learning that integrates model-conditioned failure diagnosis, retrieval-grounded data construction, supervised adaptation, and re-evaluation. Among 15 evaluated configurations, GPT-Image-2 leads single-image editing with an Overall score of 4.076, while Seedream-5.0-Pro leads text-intensive editing at 4.291. Evolution experiments across four generation and editing benchmarks demonstrate substantial targeted gains: on their report-selected Top-5 single-image targets, FLUX.2 [dev] improves Overall from 1.083 to 3.517, while Qwen-Image-Edit-2511 gains 13.2%. Together, our benchmark and harness provide a foundation for developing more reliable image models through interpretable evaluation and feedback-driven learning.

cs.CV↗

arXiv2026

WeVisOne-VL: Unified Visual Perception from Objects to Pixels

Existing vision-language models (VLMs) remain unreliable for fine-grained visual perception. In this report, we present WeVisOne-VL (WeChat Vision All-in-One), a powerful VLM that unifies object-level and pixel-level perception through structured next-token prediction. It supports multiple visual tasks including object detection, referring localization, counting, OCR, segmentation, and matting, enabling localization at any user-specific granularity, from points and boxes to masks and alpha mattes. Architecturally, we integrate a lightweight pixel decoder into the VLM via a special token to generate instance-level pixel outputs. We train the model with a three-stage curriculum SFT followed by mixed-task RL. Across over 30 benchmarks, WeVisOne-VL demonstrates leading performance on both object-level and pixel-level tasks, while maintaining robust rejection for object-absent queries. By releasing WeVisOne-VL to the open-source community, we hope to advance VLMs from broad visual understanding toward truly reliable perception systems for real-world deployment.

cs.CV↗

arXiv2026

Convergence of Markovian Iteration for $Z$-Coupled FBSDEs via Gaussian Smoothing

In this paper, we investigate the Markovian iteration method for coupled forward-backward stochastic differential equations (FBSDEs) with drift $b(t,X_t,Y_t,Z_t)$ and deterministic, time-dependent diffusion. Bender and Zhang (2008) established convergence results for Markovian iteration in the $Y$-coupled setting. Extending these results to equations with $Z$-coupling presents additional challenges, particularly in obtaining uniform control of the spatial Lipschitz constants of the discrete decoupling fields across iterations and time steps. We address this difficulty by representing the discrete $Z$-field through the spatial derivative of the Gaussian-smoothed $Y$-field. This representation, combined with weak differentiation, yields uniform spatial Lipschitz bounds for both discrete decoupling fields without requiring their classical second derivatives. Under suitable contraction conditions, we prove convergence of the Markovian iteration and well-posedness of the discrete scheme. With additional regularity assumptions, we derive an overall error bound accounting for both iteration and time discretization. Numerical experiments compare three practical implementations and demonstrate the effectiveness of the derivative-based and merged approaches.

math.NA↗

arXiv2026

An Analytic COS Method for Compound Option Valuation

We develop an analytic Fourier cosine (COS) method for the valuation of compound options. By deriving closed-form expressions for the cosine coefficients at all compound stages, the proposed method eliminates the need for numerical quadrature in intermediate exercise stages while retaining the convergence properties of the underlying COS approximation. The formulation extends to multi-stage compound structures and a broader class of payoffs, and remains applicable to a wide class of stochastic models characterized by known characteristic functions, including jump-diffusion dynamics. Numerical experiments demonstrate improved computational efficiency compared with quadrature-based implementations while maintaining high accuracy. Applications to staged real-option problems further illustrate the flexibility of the method in handling nested decision structures under different uncertainty dynamics.

q-fin.CP↗

arXiv2026

Cognitive-structured Multimodal Agent for Multimodal Understanding, Generation, and Editing

Recent unified multimodal models show a single architecture can jointly perform vision/language understanding and image generation/editing. However, they repeatedly feed all historical visual and textual inputs into a shared context window, limiting long-horizon multimodal dialogue due to visual token explosion and unreliable cross-turn referencing. We propose a Cognitive-structured Multimodal Agent that externalizes visual information into an Episodic Visual Memory and selectively reactivates relevant episodes during reasoning. The agent consists of a Perceptual Abstraction Engine for structured visual abstraction, a Cognitive Retrieval Engine for cross-turn memory retrieval, and a Multimodal Executive Controller for autonomous task inference and action planning. To address the lack of turn-level retrieval supervision in existing datasets, we develop a Unified Scenario Engine that programmatically generates structured multi-turn conversations with fine-grained retrieval annotations, enabling reinforcement learning to optimize abstraction and retrieval policies. We also construct a long-horizon visual-dialogue benchmark stratified by difficulty to evaluate episodic visual recall. Our 8B agent achieves 91.4% retrieval accuracy over 20-turn sessions, surpassing 32B baselines by +8.2% while nearly halving per-turn inference time (23.1s -> 12.7s). We further present the Cognitive-structured Multimodal Agent Harness (CMA-Harness), a tool-augmented deployment of the same cognitive structure integrating persistent multimodal memory, web access, image generation/editing/composition tools, and OpenAI-compatible serving. Structured memory and modular decision-making offer a more scalable, efficient paradigm for long-horizon multimodal agents than monolithic parameter scaling. Code: https://github.com/caseclose/cma-harness ; Project page: https://caseclose.github.io/cma-harness/

cs.CV↗

arXiv2026

Breaking Dual Bottlenecks: Evolving Unified Multimodal Models into Self-Adaptive Interleaved Visual Reasoners

Recent unified models integrate multimodal understanding and generation within a single framework. However, an "understanding-generation gap" persists, where models can capture user intent but often fail to translate this semantic knowledge into precise pixel-level manipulation. This gap results in two bottlenecks in anything-to-image task (X2I): the attention entanglement bottleneck, where blind planning struggles with complex prompts, and the visual refinement bottleneck, where unstructured feedback fails to correct imperfections efficiently. In this paper, we propose a novel framework that empowers unified models to autonomously switch between generation strategies based on instruction complexity and model capability. To achieve this, we construct a hierarchical data pipeline that constructs execution paths across three adaptive modes: direct generation for simple cases, self-reflection for quality refinement, and multi-step planning for decomposing complex scenarios. Building on this pipeline, we contribute a high-quality dataset with over 50,000 samples and implement a two-stage training strategy comprising SFT and RL. Specifically, we design step-wise reasoning rewards to ensure logical consistency and intra-group complexity penalty to prevent redundant computational overhead. Extensive experiments demonstrate that our method outperforms existing baselines on X2I, achieving superior generation fidelity among simple-to-complex instructions. The code is released at https://github.com/WeChatCV/Interleaved_Visual_Reasoner.

cs.CV↗

arXiv2026

WinTok: A Win-Win Hybrid Tokenizer via Decomposing Visual Understanding and Generation with Transferable Tokens

Building a unified visual tokenizer is essential for bridging the gap between visual understanding and generation. Yet existing approaches struggle with the inherent conflict between these tasks, as a single token space is forced to support both high-level semantic abstraction and low-level pixel reconstruction. We propose WinTok, a concise hybrid tokenizer that achieves a win-win performance by explicitly decoupling the two objectives. WinTok supplements pixel tokens with a set of learnable semantic tokens, effectively mitigating cross-task interference without incurring the computational overhead of dual tokenizers. To further enhance understanding capability, we introduce an asymmetric token distillation mechanism: the semantic tokens are guided by pretrained semantic embeddings from any visual foundation model, enabling them to inherit strong discriminative power while maintaining flexibility. Across 10 challenging benchmarks, WinTok delivers consistent improvements in reconstruction, understanding, and generation. Trained on only 50M open-source data, WinTok surpasses the strong baseline UniTok by 11.2% in classification accuracy and achieves a competitive reconstruction rFID of 0.41, despite using substantially less training data. Code is released at https://github.com/markywg/WinTok.

cs.CV↗

arXiv2026

To understand the radiative processes of pulsars and fast radio bursts with the FAST

The radiative mechanism of coherent radio emission has remained an enigma since the discovery of pulsars, even the emergence of fast radio bursts (FRBs), which exhibit similarities to the single-pulse behavior of pulsars and have opened a new view for deciphering the long-standing mystery. Besides tremendous efforts in modelling, advanced facilities matter for solving the problem. The authors review the observational breakthroughs from the Five-hundred-meter Aperture Spherical radio Telescope (FAST), which are providing pivotal insights to unravel the underlying physics of pulsars and FRBs. This study offers a novel perspective in the era when pulsars meet FRBs, and further investigations are encouraged to utilize the highly sensitive telescope, the FAST.

astro-ph.HE↗

arXiv2026

The Interstellar Scintillation of the Radio-Loud Magnetar XTE J1810-197

We present a comprehensive interstellar scintillation (ISS) study of the radio-loud magnetar XTE~J1810$-$197, based on six years of multi-frequency monitoring (2018$-$2024) with the Shanghai Tian Ma Radio Telescope (TMRT) at 7.0, 8.6, and 14.0~GHz. The scintillation parameters--decorrelation bandwidth $Δν_{\rm d}$, decorrelation time $Δτ_{\rm d}$, and drift rate $dt/dν$--are fully characterized. Our measured $Δτ_{\rm d}$ implies $Δτ_{\rm d} < 4$~s at 575-725~MHz under a Kolmogorov spectrum, which is shorter than the magnetar's 5.54~s spin period. This result naturally explains the previously reported absence of pulse-to-pulse coherence at these frequencies. Kinematic modeling locates the dominant scattering screen at $1.6\pm0.1$~kpc away from the Earth, within the Sagittarius Arm. The screen coincides with the HII region JCMTSE~J180921.2$-$201932 and is unrelated to the magnetar's 2018 outburst suggested by earlier studies. A scintillation arc detected at 14.0~GHz represents the highest-frequency arc observed to date. The asymmetry of arcs is linearly correlated with a dispersion-measure gradient across the screen ($r = 0.959$, $p < 10^{-8}$). We also measure its refractive scintillation timescale, which is only $1.21\pm0.19$~d. Clear DISS at 14~GHz effectively resolves the debate over a possible strong-to-weak scattering transition at this frequency. These results extend the ISS characterization of magnetars to previously unexplored frequencies and provide a precise probe of the ionized interstellar medium in the Sagittarius Arm.

astro-ph.HE↗

arXiv2026

WeTok: Powerful Discrete Tokenization for High-Fidelity Visual Reconstruction

Visual tokenizer is a critical component for vision generation. However, the existing tokenizers often face unsatisfactory trade-off between compression ratios and reconstruction fidelity. To fill this gap, we introduce a powerful and concise WeTok tokenizer, which surpasses the previous leading tokenizers via two core innovations. (1) Group-wise lookup-free Quantization (GQ). We partition the latent features into groups, and perform lookup-free quantization for each group. As a result, GQ can efficiently overcome memory and computation limitations of prior tokenizers, while achieving a reconstruction breakthrough with more scalable codebooks. (2) Generative Decoder (GD). Different from prior tokenizers, we introduce a generative decoder with a prior of extra noise variable. In this case, GD can probabilistically model the distribution of visual data conditioned on discrete tokens, allowing WeTok to reconstruct visual details, especially at high compression ratio. On the ImageNet 50k validation set, at a high-fidelity setting, WeTok achieves a record-low zero-shot rFID of 0.12, outperforming leading continuous tokenizers like FLUX-VAE (0.18) and SD-VAE 3.5 (0.19) with 400% compression ratio. Furthermore, in a high-compression regime, WeTok achieves a zero-shot rFID of 3.49 at a 768$\times$ compression ratio, substantially surpassing Cosmos, which scores 4.57 at only 50% our compression ratio. Code and models are available: https://github.com/zhuangshaobin/WeTok.

cs.CV↗

arXiv2026

LIT-LVM: Structured Regularization for Interaction Terms in Linear Predictors using Latent Variable Models

Some of the simplest, yet most frequently used predictors in statistics and machine learning use weighted linear combinations of features. Such linear predictors can model non-linear relationships between features by adding interaction terms corresponding to the products of all pairs of features. We consider the problem of accurately estimating coefficients for interaction terms in linear predictors. We hypothesize that the coefficients for different interaction terms have an approximate low-dimensional structure and represent each feature by a latent vector in a low-dimensional space. This low-dimensional representation can be viewed as a structured regularization approach that further mitigates overfitting in high-dimensional settings beyond standard regularizers such as the lasso and elastic net. We demonstrate that our approach, called LIT-LVM, achieves superior prediction accuracy compared to the elastic net, hierarchical lasso, and factorization machines on a wide variety of simulated and real data, particularly when the number of interaction terms is high compared to the number of samples. LIT-LVM also provides low-dimensional latent representations for features that are useful for visualizing and analyzing their relationships.

cs.LG↗

arXiv2026

The Compound BSDE Method: A Fully Forward Method for Option Pricing and Optimal Stopping Problems in Finance

We propose the Compound BSDE method, a fully forward, deep-learning-based approach for solving a broad class of problems in financial mathematics, including optimal stopping. The method is based on a reformulation of option pricing problems in terms of a system of backward stochastic differential equations (BSDEs), which offers a new perspective on the numerical treatment of compound options and optimal stopping problems such as Bermudan option pricing. Building on the classical deep BSDE method for a single BSDE, we develop an algorithm for compound BSDEs and establish its convergence properties. In particular, we derive an a posteriori error estimate for the proposed method. Numerical experiments demonstrate the accuracy and computational efficiency of the approach, and illustrate its effectiveness for high-dimensional option pricing and optimal stopping problems.

q-fin.CP↗

arXiv2026

Does the radio-active phase of XTE~J1810$-$197 recur following the same evolutionary pattern?

Magnetars are the most strongly magnetized compact objects known in the Universe and are regarded as one of the primary engines powering a variety of enigmatic, high-energy transients. However, our understanding of magnetars remains highly limited, constrained by observational sample size and radiative variability. XTE~J1810$-$197, which re-entered a radio-active phase in 2018, is one of only six known radio-pulsating magnetars. Leveraging the distinctive capability for simultaneous dual-frequency observations, we utilized the Shanghai Tianma Radio Telescope (TMRT) to monitor this magnetar continuously at both 2.25 and 8.60~GHz, capturing its entire evolution from radio activation to quenching. This enabled precise characterization of the evolution in its integrated profile, spin frequency, flux density, and spectral index ($α$, defined by $S \propto f^α$). The first time derivative of its spin frequency $\dotν$ passed through four distinct phases -- rapid decrease, violent oscillation, steady decline, and stable recovery -- before returning to its pre-outburst value concomitant with the cessation of radio emission. Remarkably, both the amplitudes and the characteristic time-scales of these $\dotν$ variations match those observed during the previous outburst that began in 2003, providing the first demonstration that post-outburst rotational evolution and radiative behavior in a magnetar are repeatable. A twisted-magnetosphere model can qualitatively account for this repeatability as well as for the progressive narrowing and abrupt disappearance of the radio pulse radiation, thereby receiving strong observational support.

astro-ph.HE↗

arXiv2025

UnicEdit-10M: A Dataset and Benchmark Breaking the Scale-Quality Barrier via Unified Verification for Reasoning-Enriched Edits

With the rapid advances of powerful multimodal models such as GPT-4o, Nano Banana, and Seedream 4.0 in Image Editing, the performance gap between closed-source and open-source models is widening, primarily due to the scarcity of large-scale, high-quality training data and comprehensive benchmarks capable of diagnosing model weaknesses across diverse editing behaviors. Existing data construction methods face a scale-quality trade-off: human annotations are high-quality but not scalable, while automated pipelines suffer from error propagation and noise. To address this, we introduce a lightweight data pipeline that replaces multi-toolchains with an end-to-end model and a unified post-verification stage. For scalable quality control, we train a 7B dual-task expert model, \textbf{Qwen-Verify}, for efficient failure detection and instruction recaptioning. This pipeline yields \textbf{UnicEdit-10M}, a 10M-scale dataset spanning diverse basic and complex editing tasks. We also propose \textbf{UnicBench}, a general benchmark that extends beyond basic edits to explicitly assess spatial and knowledge-driven reasoning. To enable fine-grained diagnosis, we introduce novel metrics, including \textit{Non-edit Consistency} and \textit{Reasoning Accuracy}. Our analysis of mainstream models on UnicBench reveals their limitations and provides clear directions for future research.

cs.CV↗

arXiv2025

Femtosecond photo-induced displacive phase transition in Sb$_{2}$Te (group 2) phase-change material

Two classes of Phase Change Materials (PCMs) have emerged as the best candidates for applications requiring the fast reading and writing of data: GeTe-Sb$_{2}$Te$_{3}$ pseudobinary alloys (group 1) and doped Sb-Te compounds near the eutectic composition Sb$_{70}$Te$_{30}$ (group 2). Both material classes undergo reversible switching between a low-resistance opaque crystalline phase and a high-resistance but less absorbing amorphous phase through heating, electrical, or optical pulses, achieving (sub-)nanosecond switching speeds. While group 1 compounds are employed in current generation devices and relatively well studied, model systems in group 2 compounds have been found to crystallize more rapidly and thus offer the perspective of improved devices. Despite their superior crystallization speed (SET process), to this point there have been no ultrafast experimental studies on crystallized PCMs of group 2 for the RESET process. Here we perform ultrafast electron diffraction and femtosecond resolved sum frequency non-linear spectroscopy on Peierls distorted Sb$_{2}$Te crystallized thin films (PCM of group 2) following femtosecond optical pulse irradiation. We observe a pump-induced structural change on two distinct timescales: responses with characteristic timescales of $\approx$ 300 fs and 2~ps. We quantified the experimental result by a coherent displacement and the Debye-Waller effect. In particular, the $\approx$ 300 fs UED signal results from the ultrafast release of the Peierls distortion through non-thermal coherent Sb displacement, while the 2~ps response reflects electron-lattice equilibrium. These results reveal the ultrafast non-thermal structural dynamics of Sb$_{2}$Te and suggest energy-efficient switching of group 2 PCMs should be possible on femtosecond time scales.

cond-mat.mtrl-sci↗

arXiv2025

Text-guided Visual Prompt DINO for Generic Segmentation

Recent advancements in multimodal vision models have highlighted limitations in late-stage feature fusion and suboptimal query selection for hybrid prompts open-world segmentation, alongside constraints from caption-derived vocabularies. To address these challenges, we propose Prompt-DINO, a text-guided visual Prompt DINO framework featuring three key innovations. First, we introduce an early fusion mechanism that unifies text/visual prompts and backbone features at the initial encoding stage, enabling deeper cross-modal interactions to resolve semantic ambiguities. Second, we design order-aligned query selection for DETR-based architectures, explicitly optimizing the structural alignment between text and visual queries during decoding to enhance semantic-spatial consistency. Third, we develop a generative data engine powered by the Recognize Anything via Prompting (RAP) model, which synthesizes 0.5B diverse training instances through a dual-path cross-verification pipeline, reducing label noise by 80.5% compared to conventional approaches. Extensive experiments demonstrate that Prompt-DINO achieves state-of-the-art performance on open-world detection benchmarks while significantly expanding semantic coverage beyond fixed-vocabulary constraints. Our work establishes a new paradigm for scalable multimodal detection and data generation in open-world scenarios. Data&Code are available at https://github.com/WeChatCV/WeVisionOne.

cs.CV↗

arXiv2025

Video-GPT via Next Clip Diffusion

GPT has shown its remarkable success in natural language processing. However, the language sequence is not sufficient to describe spatial-temporal details in the visual world. Alternatively, the video sequence is good at capturing such details. Motivated by this fact, we propose a concise Video-GPT in this paper by treating video as new language for visual world modeling. By analogy to next token prediction in GPT, we introduce a novel next clip diffusion paradigm for pretraining Video-GPT. Different from the previous works, this distinct paradigm allows Video-GPT to tackle both short-term generation and long-term prediction, by autoregressively denoising the noisy clip according to the clean clips in the history. Extensive experiments show our Video-GPT achieves the state-of-the-art performance on video prediction, which is the key factor towards world modeling (Physics-IQ Benchmark: Video-GPT 34.97 vs. Kling 23.64 vs. Wan 20.89). Moreover, it can be well adapted on 6 mainstream video tasks in both video generation and understanding, showing its great generalization capacity in downstream. The project page is at https://zhuangshaobin.github.io/Video-GPT.github.io/.

cs.CV↗