arXiv subjects

Gang Cheng

Publications and source records attributed to Gang Cheng.

14 recordsLinked to original sources

arXiv2026

Wan-Animate-2: Pushing the Application Boundaries of Character Animation

Character image animation remains a foundational yet challenging task in computer vision. Existing approaches can be broadly categorized into three paradigms: methods based on explicit motion representations suffer from extraction errors and identity drift; methods based on implicit motion features lose fine-grained dynamics through compression; and in-context learning approaches avoid intermediate representations but incur prohibitive computational costs. Furthermore, all current systems are designed for offline synthesis, unable to meet the real-time requirements of interactive applications such as digital avatars and live-streaming hosts. To address these limitations, we present Wan-Animate-2, an end-to-end character animation framework that directly consumes the driving video within a redesigned Diffusion Transformer. Our architecture achieves superior motion fidelity and identity preservation by eliminating intermediate motion extractors entirely. We further introduce text driven viewpoint control that decouples the output camera perspective from the driving video--a capability rarely supported by prior character animation methods that rely on explicit motion representations. Beyond generation quality, we present Wan-Animate-2-Lite, an efficient variant that reduces inference latency to real-time thresholds through a three-stage training paradigm: teacher forcing pretraining with error buffer mechanism, and Self-Forcing distillation with chunk-wise backpropagation. This enables streaming character animation for interactive applications, opening new deployment scenarios that were previously infeasible. Qualitative evaluations and user studies demonstrate that Wan-Animate-2 achieves high-fidelity animation results across diverse characters and motion patterns. To foster further research and community development, we will release the Wan-Animate-2-Base model weights to the public.

cs.CV↗

arXiv2026

Wan-Dancer: A Hierarchical Framework for Minute-scale Coherent Music-to-Dance Generation

Generating long-duration, high-definition, and rhythmically synchronized dance videos directly from music remains a significant challenge, primarily due to the temporal constraints of current diffusion models, which typically fail beyond 20 seconds. Existing approaches, whether they rely on intermediate 3D skeletons or on end-to-end video synthesis, suffer from temporal drift, identity inconsistency, and repetitive motion patterns when extended to longer horizons. To address these limitations, we propose a novel hierarchical framework for minute-scale coherent music-to-dance generation. Our method decouples the process into global keyframe planning and local temporal refinement, leveraging full-track musical context to ensure long-range coherence. Key innovations include dynamic frame rate adaptation via time-mapped RoPE embeddings for precise alignment, an optical-flow-based loss function to enhance motion continuity, and motion-speed control to preserve high-fidelity details during rapid movements. Extensive experiments demonstrate that our framework surpasses the conventional duration barrier, generating stable, 720p/30fps videos exceeding one minute with superior temporal stability. Furthermore, the model exhibits robust versatility across five distinct dance genres, conditioned on both audio and textual prompts, establishing a new state-of-the-art in coherent, long-form dance video synthesis.

cs.CV↗

arXiv2025

Wan-Animate: Unified Character Animation and Replacement with Holistic Replication

We introduce Wan-Animate, a unified framework for character animation and replacement. Given a character image and a reference video, Wan-Animate can animate the character by precisely replicating the expressions and movements of the character in the video to generate high-fidelity character videos. Alternatively, it can integrate the animated character into the reference video to replace the original character, replicating the scene's lighting and color tone to achieve seamless environmental integration. Wan-Animate is built upon the Wan model. To adapt it for character animation tasks, we employ a modified input paradigm to differentiate between reference conditions and regions for generation. This design unifies multiple tasks into a common symbolic representation. We use spatially-aligned skeleton signals to replicate body motion and implicit facial features extracted from source images to reenact expressions, enabling the generation of character videos with high controllability and expressiveness. Furthermore, to enhance environmental integration during character replacement, we develop an auxiliary Relighting LoRA. This module preserves the character's appearance consistency while applying the appropriate environmental lighting and color tone. Experimental results demonstrate that Wan-Animate achieves state-of-the-art performance. We are committed to open-sourcing the model weights and its source code.

cs.CV↗

arXiv2025

PanoLAM: Large Avatar Model for Gaussian Full-Head Synthesis from One-shot Unposed Image

We present a feed-forward framework for Gaussian full-head synthesis from a single unposed image. Unlike previous work that relies on time-consuming GAN inversion and test-time optimization, our framework can reconstruct the Gaussian full-head model given a single unposed image in a single forward pass. This enables fast reconstruction and rendering during inference. To mitigate the lack of large-scale 3D head assets, we propose a large-scale synthetic dataset from trained 3D GANs and train our framework using only synthetic data. For efficient high-fidelity generation, we introduce a coarse-to-fine Gaussian head generation pipeline, where sparse points from the FLAME model interact with the image features by transformer blocks for feature extraction and coarse shape reconstruction, which are then densified for high-fidelity reconstruction. To fully leverage the prior knowledge residing in pretrained 3D GANs for effective reconstruction, we propose a dual-branch framework that effectively aggregates the structured spherical triplane feature and unstructured point-based features for more effective Gaussian head reconstruction. Experimental results show the effectiveness of our framework towards existing work. Project page at: https://panolam.github.io/.

cs.CV↗

arXiv2025

Learning to Conceal Risk: Controllable Multi-turn Red Teaming for LLMs in the Financial Domain

Large Language Models (LLMs) are increasingly deployed in finance, where unsafe behavior can lead to serious regulatory risks. However, most red-teaming research focuses on overtly harmful content and overlooks attacks that appear legitimate on the surface yet induce regulatory-violating responses. We address this gap by introducing a controllable black-box multi-turn risk-concealed red-teaming framework (CoRT) that progressively conceals surface-level risk while exploiting regulatory-violating behaviors. CoRT contains two key components: (i) a Risk Concealment Attacker (RCA) that generates multi-turn prompts via iterative refinement, and (ii) a Risk Concealment Controller (RCC) that predicts a turn-level Risk Concealment Score (RCS) to steer RCA's follow-up style. We also built a domain-specific benchmark, FinRisk-Bench, with 522 instructions spanning six financial risk categories. Experiments on nine widely used LLMs show that CoRT (RCA) achieves 93.19% average attack success rate (ASR), and CoRT (RCA+RCC) further improves the average ASR to 95.00%. Our code and FinRisk-Bench are available at https://github.com/gcheng128/CoRT.

cs.CL↗

arXiv2024

Modeling Analog Dynamic Range Compressors using Deep Learning and State-space Models

We describe a novel approach for developing realistic digital models of dynamic range compressors for digital audio production by analyzing their analog prototypes. While realistic digital dynamic compressors are potentially useful for many applications, the design process is challenging because the compressors operate nonlinearly over long time scales. Our approach is based on the structured state space sequence model (S4), as implementing the state-space model (SSM) has proven to be efficient at learning long-range dependencies and is promising for modeling dynamic range compressors. We present in this paper a deep learning model with S4 layers to model the Teletronix LA-2A analog dynamic range compressor. The model is causal, executes efficiently in real time, and achieves roughly the same quality as previous deep-learning models but with fewer parameters.

cs.SD↗

arXiv2023

Modeling of SCADA and PMU Measurement Chains

In this document, the supervisory control and data acquisition (SCADA) and phasor measurement unit (PMU) measurement chain modeling will be studied, where the measurement error sources of each component in the SCADA and PMU measurement chains and the reasons leading to measurement errors exhibiting non-zero-mean, non-Gaussian, and time-varying statistical characteristic are summarized and analyzed. This document provides a few equations, figures, and discussions about the details of the SCADA and PMU measurement error chain modeling, which are intended to facilitate the understanding of how the measurement errors are designed for each component in the SCADA and PMU measurement chains. The measurement chain models described here are also used for synthesizing measurement errors with realistic characteristics in simulation cases to test the developed algorithms or methodologies.

eess.SP↗

arXiv2022

Handling Nonmonotone Missing Data with Available Complete-Case Missing Value Assumption

Nonmonotone missing data is a common problem in scientific studies. The conventional ignorability and missing-at-random (MAR) conditions are unlikely to hold for nonmonotone missing data and data analysis can be very challenging with few complete data. In this paper, we introduce the available complete-case missing value (ACCMV) assumption for handling nonmonotone and missing-not-at-random (MNAR) problems. Our ACCMV assumption is applicable to data set with a small set of complete observations and we show that the ACCMV assumption leads to nonparametric identification of the distribution for the variables of interest. We further propose an inverse probability weighting estimator, a regression adjustment estimator, and a multiply-robust estimator for estimating a parameter of interest. We studied the underlying asymptotic and efficiency theories of the proposed estimators. We show the validity of our method with simulation studies and further illustrate the applicability of our method by applying it to a diabetes data set from electronic health records.

stat.ME↗

arXiv2022

Long-term effect estimation when combining clinical trial and observational follow-up datasets

Combining experimental and observational follow-up datasets has received a lot of attention lately. In a time-to-event setting, recent work has used medicare claims to extend the follow-up period for participants in a prostate cancer clinical trial. This allows the estimation of the long-term effect that cannot be estimated by clinical trial data alone. In this paper, we study the estimation of long-term effect when participants in a clinical trial are linked to an observational follow-up dataset with incomplete data. Such data linkages are often incomplete for various reasons. We formulate incomplete linkages as a missing data problem with careful considerations of the relationship between the linkage status and the missing data mechanism. We use the popular Cox proportional hazard model as a working model to define the long-term effect. We propose a conditional linking at random (CLAR) assumption and an inverse probability of linkage weighting (IPLW) partial likelihood estimator. We show that our IPLW partial likelihood estimator is consistent and asymptotically normal. We further extend our approach to incorporate time-dependent covariates. Simulations results confirm the validity of our method, and we further apply our methods to the SWOG study.

stat.ME↗

arXiv2017

Nonparametric Inference via Bootstrapping the Debiased Estimator

In this paper, we propose to construct confidence bands by bootstrapping the debiased kernel density estimator (for density estimation) and the debiased local polynomial regression estimator (for regression analysis). The idea of using a debiased estimator was recently employed by Calonico et al. (2018b) to construct a confidence interval of the density function (and regression function) at a given point by explicitly estimating stochastic variations. We extend their ideas of using the debiased estimator and further propose a bootstrap approach for constructing simultaneous confidence bands. This modified method has an advantage that we can easily choose the smoothing bandwidth from conventional bandwidth selectors and the confidence band will be asymptotically valid. We prove the validity of the bootstrap confidence band and generalize it to density level sets and inverse regression problems. Simulation studies confirm the validity of the proposed confidence bands/sets. We apply our approach to an Astronomy dataset to show its applicability

stat.ME↗

arXiv2015

Forecast Combination Under Heavy-Tailed Errors

Forecast combination has been proven to be a very important technique to obtain accurate predictions. In many applications, forecast errors exhibit heavy tail behaviors for various reasons. Unfortunately, to our knowledge, little has been done to deal with forecast combination for such situations. The familiar forecast combination methods such as simple average, least squares regression, or those based on variance-covariance of the forecasts, may perform very poorly. In this paper, we propose two nonparametric forecast combination methods to address the problem. One is specially proposed for the situations that the forecast errors are strongly believed to have heavy tails that can be modeled by a scaled Student's t-distribution; the other is designed for relatively more general situations when there is a lack of strong or consistent evidence on the tail behaviors of the forecast errors due to shortage of data and/or evolving data generating process. Adaptive risk bounds of both methods are developed. Simulations and a real example show superior performance of the new methods.

stat.ME↗

arXiv2015

On the Forecast Combination Puzzle

It is often reported in forecast combination literature that a simple average of candidate forecasts is more robust than sophisticated combining methods. This phenomenon is usually referred to as the "forecast combination puzzle". Motivated by this puzzle, we explore its possible explanations including estimation error, invalid weighting formulas and model screening. We show that existing understanding of the puzzle should be complemented by the distinction of different forecast combination scenarios known as combining for adaptation and combining for improvement. Applying combining methods without consideration of the underlying scenario can itself cause the puzzle. Based on our new understandings, both simulations and real data evaluations are conducted to illustrate the causes of the puzzle. We further propose a multi-level AFTER strategy that can integrate the strengths of different combining methods and adapt intelligently to the underlying scenario. In particular, by treating the simple average as a candidate forecast, the proposed strategy is shown to avoid the heavy cost of estimation error and, to a large extent, solve the forecast combination puzzle.

stat.ME↗

arXiv2012

Novel Photovoltaic Phenomenon in Manganite/ZnO Heterostructure

In this paper, we report a novel photovoltaic phenomenon in a low cost manganite/ZnO p-n heterojunction grown on ITO glass substrate by pulsed laser depositon (PLD) under relative low growth temperature. The heterostructure ITO/La0.62Ca0.29K0.09MnO3(LCKMO)/ZnO/Al exhibits reproducible rectifying characteristics and consists of four parts: aluminum top electrode, ITO bottom electrode, manganite and ZnO semiconductor films. Moreover, the light current can generate under continuous laser (λ=325nm) irradiation. In this article, we investigate the influence of manganite and ZnO film thickness on the electrical and photoelectric characteristics of the heterostructure at room temperature. The maximum power conversion efficiency (PCE) is achieved not only when the LCKMO and ZnO layer is thin enough, but also when the full space charge layer is sufficient. We obtain the maximum value(0.0145%) of PCE when the thickness of LCKMO and ZnO layer is 25nm and 150nm, respectively. Under this condition, the open circuit voltage is 0.04V, which can be attributed to internal photoemission.

cond-mat.mtrl-sci↗

arXiv2011

Evaluation of nano-frictional and mechanical properties of a novel Langmuir-Blodgett monolayer/self-assembly monolayer composite structure

A novel stearic acid (SA)/3-aminopropyltrethoxysilane (APS) composite structure was fabricated using the combined method of the Langmuir-Blodgett technique and self-assembly monolayer (SAM) technique. Its frictional, adhesive properties and interface contact types between the atomic force microscope tip and the samples were evaluated based on Amonton's laws and the general Carpick's transition equation, respectively. The results showed that the tip-sample contacts corresponded to the Johnson-Kendall-Robert/Derjaguin-Muller-Toporov (DMT) transition model for SiO2, APS-SAMs, and the unheated SA-APS composite structure, and for the heated SA-APS bilayer to the DMT model. Frictional forces for the four samples were linearly dependent on external loads at higher loads, and at lower loads they were significantly affected by adhesive forces. Frictional and scratching tests showed that the heated SA-APS composite structure exhibited the best lubricating properties and adhesion resistance ability, and its wear resistance capacity was greatly improved due to the binding-mode conversion from hydrogen bonds to covalent bonds. Thus, this kind of composite bilayer might be promising for applications in the lubrication of nano/microelectromechanical systems. I.

cond-mat.mtrl-sci↗