arXiv subjects

Zhichao Wu

Publications and source records attributed to Zhichao Wu.

14 recordsLinked to original sources

arXiv2026

Prioritized Rollouts for Efficient World Model-based Vision-Language-Action Policy Optimization

Vision-Language-Action (VLA) models have emerged as a powerful paradigm for embodied intelligence, but fine-tuning them with reinforcement learning (RL) remains constrained by the cost of real-world robot interaction. Model-based reinforcement learning (MBRL) reduces this cost by using a learned world model to generate rollouts for policy optimization. However, it becomes computationally expensive as VLA policies and world models scale. Existing methods typically treat states equally, overlooking substantial differences in their utility for policy improvement. In this paper, we show that policy uncertainty helps identify states with greater potential for policy improvement. The policy exhibits high uncertainty at only a small subset of states, often during decision-sensitive stages where small action differences can alter task outcomes, suggesting that policy improvements at these states could be particularly valuable. Building on these findings, we introduce U-GROW, a lightweight, plug-and-play sampling layer that directs more model rollouts to these informative states. By modifying only the branched-start distribution, U-GROW can be integrated into existing MBRL pipelines without changing the policy optimization objective. Experiments in both simulated and real-world manipulation tasks demonstrate the efficiency and effectiveness of U-GROW, supporting the use of policy uncertainty to guide experience generation.

cs.LG↗

arXiv2026

VINE: Taming Generative Control Policies for Reinforcement Learning

Flow-matching policies have emerged as an effective policy parameterization for robot learning. They iteratively generate actions from noise, enabling highly expressive modeling of complex and multimodal action distributions. However, prior works observed that scaling these policies with value-gradient reinforcement learning (RL) often leads to training instability. Existing methods attribute this instability to iterative generation and therefore avoid end-to-end value-gradient optimization by sacrificing iterative generation, high expressiveness, or value-gradient optimization. Contrary to prior belief, we show the instability does not stem from iterative generation itself, but from the vanilla sampling strategy originally designed for behavior cloning, which becomes brittle under value-gradient RL. Motivated by this insight, we propose VINE, an RL-oriented sampling method that enables stable end-to-end value-gradient optimization for flow-matching policies. Instead of following a single flow trajectory, VINE reconstructs a new interpolation state at every denoising step, creating a stable differentiable path for value-gradient propagation while remaining compatible with the original flow-matching denoising process. As a result, VINE preserves the expressiveness and iterative generation of flow-matching without sacrificing end-to-end value-gradient optimization. Despite performing end-to-end backpropagation through all ten denoising steps, VINE achieves stable policy improvement and consistently outperforms state-of-the-art RL methods on the OGBench offline RL benchmark and real-world robotic manipulation task. Videos are available on our website: https://agibottech.github.io/vine.

cs.RO↗

arXiv2026

ALOE: Action-Level Off-Policy Evaluation for Vision-Language-Action Model Post-Training

We study how to improve large foundation vision-language-action (VLA) systems through human-in-the-loop reinforcement learning (RL) in real-world environments. A key challenge is learning reliable value functions from heterogeneous real-world experience, as value estimation provides the primary learning signal for VLA training. In practice, replay buffers contain trajectories collected from historical policies, online rollouts, demonstrations, and intermittent human interventions. Because replay buffers mix trajectories generated by different behaviors, the observed returns can be mismatched with the quality of the current policy. Prior VLA post-training methods often rely on progress-style value signals, which reflect the average quality of historical behaviors, leading to mismatched learning signals for the current policy. In this paper, we propose ALOE, an off-policy evaluation framework whose value function directly evaluates current-policy behavior for each iteration. Specifically, ALOE combines chunked temporal-difference bootstrapping and conservative value aggregation to perform stable current-policy evaluation, then uses these estimates for advantage-weighted policy improvement. This design improves credit assignment to critical action chunks under sparse rewards and supports stable policy improvement. We evaluate ALOE on four real-world manipulation tasks encompassing long-horizon and high-precision scenarios: smartphone packing, laundry folding, multi-object sorting, and phone assembly. Across all tasks, ALOE outperforms other VLA post-training methods, highlighting the benefit of off-policy value estimates for real-world VLA post-training. Videos are available at our project website https://rooshy-yang.github.io/aloe.

cs.RO↗

arXiv2026

Towards Practical World Model-based Reinforcement Learning for Vision-Language-Action Models

Vision-Language-Action (VLA) models show strong generalization for robotic control, but finetuning them with reinforcement learning (RL) is constrained by the high cost and safety risks of real-world interaction. Training VLA models in interactive world models avoids these issues but introduces several challenges, including pixel-level world modeling, multi-view consistency, and compounding errors under sparse rewards. Building on recent advances across large multimodal models and model-based RL, we propose VLA-MBPO, a practical framework to tackle these problems in VLA finetuning. Our approach has three key design choices: (i) adapting unified multimodal models (UMMs) for data-efficient world modeling; (ii) an interleaved view decoding mechanism to enforce multi-view consistency; and (iii) chunk-level branched rollout to mitigate error compounding. Theoretical analysis and experiments across simulation and real-world tasks demonstrate that VLA-MBPO significantly improves policy performance and sample efficiency, underscoring its robustness and scalability for real-world robotic deployment.

cs.RO↗

arXiv2026

Speedup Patch: Learning a Plug-and-Play Policy to Accelerate Embodied Manipulation

While current embodied policies exhibit remarkable manipulation skills, their execution remains unsatisfactorily slow as they inherit the tardy pacing of human demonstrations. Existing acceleration methods typically require policy retraining or costly online interactions, limiting their scalability for large-scale foundation models. In this paper, we propose Speedup Patch (SuP), a lightweight, policy-agnostic framework that enables plug-and-play acceleration using solely offline data. SuP introduces an external scheduler that adaptively downsamples action chunks provided by embodied policies to eliminate redundancies. Specifically, we formalize the optimization of our scheduler as a Constrained Markov Decision Process (CMDP) aimed at maximizing efficiency without compromising task performance. Since direct success evaluation is infeasible in offline settings, SuP introduces World Model based state deviation as a surrogate metric to enforce safety constraints. By leveraging a learned world model as a virtual evaluator to predict counterfactual trajectories, the scheduler can be optimized via offline reinforcement learning. Empirical results on simulation benchmarks (Libero, Bigym) and real-world tasks validate that SuP achieves an overall 1.8x execution speedup for diverse policies while maintaining their original success rates.

cs.RO↗

arXiv2025

MPE-TTS: Customized Emotion Zero-Shot Text-To-Speech Using Multi-Modal Prompt

Most existing Zero-Shot Text-To-Speech(ZS-TTS) systems generate the unseen speech based on single prompt, such as reference speech or text descriptions, which limits their flexibility. We propose a customized emotion ZS-TTS system based on multi-modal prompt. The system disentangles speech into the content, timbre, emotion and prosody, allowing emotion prompts to be provided as text, image or speech. To extract emotion information from different prompts, we propose a multi-modal prompt emotion encoder. Additionally, we introduce an prosody predictor to fit the distribution of prosody and propose an emotion consistency loss to preserve emotion information in the predicted prosody. A diffusion-based acoustic model is employed to generate the target mel-spectrogram. Both objective and subjective experiments demonstrate that our system outperforms existing systems in terms of naturalness and similarity. The samples are available at https://mpetts-demo.github.io/mpetts_demo/.

cs.SD↗

arXiv2023

DCTTS: Discrete Diffusion Model with Contrastive Learning for Text-to-speech Generation

In the Text-to-speech(TTS) task, the latent diffusion model has excellent fidelity and generalization, but its expensive resource consumption and slow inference speed have always been a challenging. This paper proposes Discrete Diffusion Model with Contrastive Learning for Text-to-Speech Generation(DCTTS). The following contributions are made by DCTTS: 1) The TTS diffusion model based on discrete space significantly lowers the computational consumption of the diffusion model and improves sampling speed; 2) The contrastive learning method based on discrete space is used to enhance the alignment connection between speech and text and improve sampling quality; and 3) It uses an efficient text encoder to simplify the model's parameters and increase computational efficiency. The experimental results demonstrate that the approach proposed in this paper has outstanding speech synthesis quality and sampling speed while significantly reducing the resource consumption of diffusion model. The synthesized samples are available at https://github.com/lawtherWu/DCTTS.

cs.SD↗

arXiv2022

A deep learning framework for the detection and quantification of drusen and reticular pseudodrusen on optical coherence tomography

Purpose - To develop and validate a deep learning (DL) framework for the detection and quantification of drusen and reticular pseudodrusen (RPD) on optical coherence tomography scans. Design - Development and validation of deep learning models for classification and feature segmentation. Methods - A DL framework was developed consisting of a classification model and an out-of-distribution (OOD) detection model for the identification of ungradable scans; a classification model to identify scans with drusen or RPD; and an image segmentation model to independently segment lesions as RPD or drusen. Data were obtained from 1284 participants in the UK Biobank (UKBB) with a self-reported diagnosis of age-related macular degeneration (AMD) and 250 UKBB controls. Drusen and RPD were manually delineated by five retina specialists. The main outcome measures were sensitivity, specificity, area under the ROC curve (AUC), kappa, accuracy and intraclass correlation coefficient (ICC). Results - The classification models performed strongly at their respective tasks (0.95, 0.93, and 0.99 AUC, respectively, for the ungradable scans classifier, the OOD model, and the drusen and RPD classification model). The mean ICC for drusen and RPD area vs. graders was 0.74 and 0.61, respectively, compared with 0.69 and 0.68 for intergrader agreement. FROC curves showed that the model's sensitivity was close to human performance. Conclusions - The models achieved high classification and segmentation performance, similar to human performance. Application of this robust framework will further our understanding of RPD as a separate entity from drusen in both research and clinical settings.

eess.IV↗

arXiv2020

Unsupervised Image Segmentation using Mutual Mean-Teaching

Unsupervised image segmentation aims at assigning the pixels with similar feature into a same cluster without annotation, which is an important task in computer vision. Due to lack of prior knowledge, most of existing model usually need to be trained several times to obtain suitable results. To address this problem, we propose an unsupervised image segmentation model based on the Mutual Mean-Teaching (MMT) framework to produce more stable results. In addition, since the labels of pixels from two model are not matched, a label alignment algorithm based on the Hungarian algorithm is proposed to match the cluster labels. Experimental results demonstrate that the proposed model is able to segment various types of images and achieves better performance than the existing methods.

cs.CV↗

arXiv2019

Modeling Disease Progression In Retinal OCTs With Longitudinal Self-Supervised Learning

Longitudinal imaging is capable of capturing the static ana\-to\-mi\-cal structures and the dynamic changes of the morphology resulting from aging or disease progression. Self-supervised learning allows to learn new representation from available large unlabelled data without any expert knowledge. We propose a deep learning self-supervised approach to model disease progression from longitudinal retinal optical coherence tomography (OCT). Our self-supervised model takes benefit from a generic time-related task, by learning to estimate the time interval between pairs of scans acquired from the same patient. This task is (i) easy to implement, (ii) allows to use irregularly sampled data, (iii) is tolerant to poor registration, and (iv) does not rely on additional annotations. This novel method learns a representation that focuses on progression specific information only, which can be transferred to other types of longitudinal problems. We transfer the learnt representation to a clinically highly relevant task of predicting the onset of an advanced stage of age-related macular degeneration within a given time interval based on a single OCT scan. The boost in prediction accuracy, in comparison to a network learned from scratch or transferred from traditional tasks, demonstrates that our pretrained self-supervised representation learns a clinically meaningful information.

eess.IV↗

arXiv2016

Group velocity locked vector dissipative solitons in a high repetition rate fiber laser

Vectorial nature of dissipative solitons (DSs) with high repetition rates is studied for the first time in a normal-dispersion fiber laser. Despite the fact that the formed DSs are strongly chirped and the repetition rate is greater than 100 MHz, polarization locked and polarization rotating group velocity locked vector DSs can be formed under 129.3 MHz fundamental mode-locking and 258.6 MHz harmonic mode-locking of the fiber laser, respectively. The two orthogonally polarized components of these vector DSs possess distinctly different central wavelengths and travel together at the same group velocity in the laser cavity, resulting in a gradual spectral edge and small steps on the optical spectra, which can be considered as an auxiliary indicator of the group velocity locked vector DSs.

physics.optics↗

arXiv2016

Dynamics of dissipative solitons in a high repetition rate normal-dispersion erbium-doped fiber laser

The dynamics of dissipative solitons (DSs) are explored in a high repetition rate normal-dispersion erbium-doped fiber laser for the first time. Despite of the high fundamental repetition rate of 129 MHz and thus the low pulse energy, a DS train with a dechirped pulse width of 418 fs, period-doubling of single and dual DSs, as well as 258 MHz 2nd-order harmonic mode-locking of DSs can be observed in the fiber laser with increasing pump power and appropriate settings. A transmitted semiconductor saturable absorber and a wavelength division multiplexer/isolator/tap hybrid module are employed to simplify the laser configuration, thus not only increasing the repetition rate, but also enhancing the stability and robustness of the fiber laser due to the commercial availability of all the components.

physics.optics↗

arXiv2016

Scalar - vector soliton fiber lasers

Rapid progress in passively mode-locked fiber lasers is currently driven by the recent discovery of vector feature of mode-locking pulses, namely, the group velocity-locked vector solitons, the phase locked vector solitons, and the high-order vector solitons. Those vector solitons are fundamentally different from the previously known scalar solitons. Here, we report a fiber laser where the mode-locked pulse evolves as a vector soliton in the strong birefringent segment and is transformed into a regular scalar soliton after the polarizer within the laser cavity. The existence of solutions in a polarization-dependent cavity comprising a periodic combination of two distinct nonlinear waves is novel and likely to be applicable to various other nonlinear systems. For very large local birefringence, our laser approaches the working regime of vector soliton lasers, while it approaches scalar soliton fiber lasers under the conditions of very small birefringence.

physics.optics↗

arXiv2015

Generation of pseudo-high-order group velocity locked vector solitons in fiber lasers

We propose and experimentally demonstrate the generation of pseudo-high-order group velocity locked vector solitons (GVLVS) in a fiber laser using a SESAM as the mode locker. With the help of an external all-fiber polarization resolved system, a GVLVS with a two-humped pulse along one polarization while a single-humped pulse along the orthogonal polarization could be obtained. The phase difference between the two humps is 180 degree.

physics.optics↗