Searcharxiv⌕ Search

arXiv subjects

Yue Wang

Publications and source records attributed to Yue Wang.

At least 37 records · Page 2Linked to original sources
arXiv2026

PathTime-VLA: Path-Time Decoupling for Factorized Post-Training of Vision-Language-Action Policies

Vision-Language-Action (VLA) policies typically predict actions at fixed time intervals, coupling the route a robot follows with its execution pace. This coupling complicates adaptation from teleoperation: useful geometric guidance comes with timing shaped by interface delays and operator behavior. Our key insight is to bring the path-time parameterization of classical motion planning into the learned action representation of a VLA. We introduce PathTime-VLA, which represents motion as a progress-indexed interaction path $X(s)$ and a positive interval-time profile. The latter defines a monotone time law $t(s)$, yielding controller commands $X(s(t))$. For a given path, alternative executions are expressed through the time profile, allowing chunk-wise speed choices without changing the geometric prediction target. This representation supports a staged post-training procedure: demonstrations and DAgger interventions establish a target-domain prior, Speed-DQN learns execution multipliers from robot interaction, and Path-AWR uses rollout outcomes to refine the diffusion path generator. A path-conditioned action expert realizes the resulting motions while maintaining distinct learning interfaces for path generation and execution timing. Across three tasks, the complete method achieves $58/60$ successes versus $57/60$ for PathTime-VLA under BC + DAgger at fixed $1\times$, with approximately $39$-$52\%$ shorter mean completion times over successful trials.

cs.RO↗
arXiv2026

GRPODropout: Less is More for Online Reinforcement Learning Rollouts

Reinforcement learning (RL) methods such as GRPO substantially improve large language model reasoning but often suffer from policy entropy collapse: the loss of sampling diversity weakens exploration and limits further improvement. Existing methods address this issue either through algorithm-level interventions, such as reward modification and entropy/KL regularization, or through token-level reweighting. We investigate a complementary perspective: entropy collapse can also be mitigated by changing which generated rollouts contribute to policy updates. Under the same sampling budget, not all rollouts contribute positively to an update, and selectively excluding some can improve learning. To address this, we propose GRPODropout: before the standard update, we use a simple strategy that selectively removes a small number of high-probability positive-advantage rollouts and recenters the retained advantages. To motivate this design, we develop a rollout-level theoretical analysis that guides method design and threshold selection. The method changes only rollout usage, and adds negligible computational overhead. Experiments show higher accuracy than original GRPO and higher actor entropy while using fewer rollout samples for updates, illustrating "less is more." This work provides insight into RL rollout usage: removing some rollouts can improve performance. Code is available at https://github.com/hexuandeng/GRPODropout/.

cs.LG↗
arXiv2026

Reflex: Visually Grounded Reactive Humanoid Control

Humans effortlessly react to the world; they adeptly return a tennis shot, catch a falling cup, step aside from a passing cart, or pass a box tossed during truck unloading. These everyday actions require anticipating a moving target and reflexively responding with the whole body within about a second. Such visual reactive whole-body control remains challenging for humanoids, which must (1) coordinate whole-body control, (2) react to object dynamics in real time, and (3) ground perception in egocentric vision. We study catching thrown boxes as a representative task that demands all three capabilities; existing systems cover only part of it, relying on end-effectors, motion capture, or marked objects. We present Reflex, a three-stage framework that learns these capabilities stage by stage while connecting them through a compact representation of object dynamics. First, a state-based policy learns whole-body catching with reinforcement learning. Second, the policy learns to infer the dynamics of a moving box from delayed and incomplete observations. Third, an egocentric RGB-D encoder learns to recover this dynamics representation directly from visual observations while keeping the whole-body controller fixed. This decomposition provides each capability with a direct training signal and makes visual learning substantially more scalable. In simulation, Reflex achieves 85.7% catch success from egocentric RGB-D, compared with 87.9% when given privileged box observations. Temporal visual history is critical: removing it reduces success by 35.6 and 39.8 points on centered and laterally offset throws. On a real Unitree G1, Reflex catches human-thrown boxes with 65% and 40% success at 2.5 and 5 m, reaching 76% and 59% of human catchers' success at the same distances.

cs.RO↗
arXiv2026

Dual-Pathway Circuits of Object Hallucination in Vision-Language Models

Vision-language models (VLMs) have demonstrated remarkable capabilities in bridging visual perception and natural language understanding, enabling a wide range of multimodal reasoning tasks. However, they often produce object hallucinations, describing content absent from the input image, which limits their reliability and interpretability. To address this limitation, we propose Dual-Pathway Circuit Analysis, a framework that identifies and characterizes hallucination-related circuits in VLMs for mechanistic understanding and causal probing. We first apply activation patching across five architecturally diverse VLMs to identify a visual grounding pathway that supports correct predictions and a hallucination pathway that drives erroneous outputs. We then introduce Conditional Pathway Analysis (CPA) to characterize pathway-level interactions, revealing that grounding components remain strongly redundant in both correct and hallucinating samples but undergo a consistent polarity flip, shifting from supporting the ground truth on correct samples to aligning with the hallucinated answer on erroneous ones. We further perform targeted suppression of hallucination-pathway components, showing that scaling these components reduces object hallucination by up to 76% with minimal accuracy cost, and validate that the same circuit selectively transfers to relational but not attribute hallucination. Evaluations on POPE-adversarial and AMBER show that the identified circuits are consistent across architectures, support causal intervention, and transfer selectively across hallucination types. Code is available at https://github.com/jiaxin26/DualPath-VLM.

cs.CV↗
arXiv2026

UniST-Pred: A Robust Unified Framework for Spatio-Temporal Traffic Forecasting in Transportation Networks Under Disruptions

Spatio-temporal traffic forecasting is a core component of intelligent transportation systems, supporting various downstream tasks such as signal control and network-level traffic management. In real-world deployments, forecasting models must operate under structural and observational uncertainties, conditions that are rarely considered in model design. Recent approaches achieve strong short-term predictive performance by tightly coupling spatial and temporal modeling, often at the cost of increased complexity and limited modularity. In contrast, efficient time-series models capture long-range temporal dependencies without relying on explicit network structure. We propose UniST-Pred, a unified spatio-temporal forecasting framework that first decouples temporal modeling from spatial representation learning, then integrates both through adaptive representation-level fusion. To assess robustness of the proposed approach, we construct a dataset based on an agent-based, microscopic traffic simulator (MATSim) and evaluate UniST-Pred under severe network disconnection scenarios. Additionally, we benchmark UniST-Pred on standard traffic prediction datasets, demonstrating its competitive performance against existing well-established models despite a lightweight design. The results illustrate that UniST-Pred maintains strong predictive performance across both real-world and simulated datasets, while also yielding interpretable spatio-temporal representations under infrastructure disruptions. The source code and the generated dataset are available at https://anonymous.4open.science/r/UniST-Pred-EF27

cs.LG↗
arXiv2026

Mechanistic transition between inter- and trans-granular creep cracking via a unified crystal plasticity-fracture framework

High-temperature creep in metals remains challenging due to the complex interplay of deformation and failure mechanisms across multiple length scales. The trade-off between inter- and trans-granular cracking governs the transition of creep rupture behaviour in metallic materials. However, the micromechanical basis of the transition and its linkage to the macroscopic brittle to ductile transition in creep rupture behaviour remain unresolved. Here, the stress-dependent competition between grain-boundary cavitation and slip-induced intragranular cracking is revealed via a unified crystal plasticity fracture framework. The framework incorporates crystal plasticity, grain-boundary cavitation, and crystallographic crack propagation within an explicit polycrystalline microstructure. The microscale computation reveals the transition from trans-granular-dominated cracking at high stress levels to mixed inter- and trans-granular cracking at intermediate stresses and inter-granular-dominated cracking at lower stresses, accompanied by distinct damage localization modes and the evolution of crack paths from grain boundary to grain-interior propagation. The results further quantify the linkage between microscopic fracture mechanisms and macroscopic creep rupture behaviour, including changes in rupture ductility and creep life. Under high-stress conditions, deformation-driven trans-granular damage localization promotes ductile rupture through extensive strain accumulation, whereas under lower-stress conditions, time-dependent grain-boundary damage accumulation leads to quasi-brittle rupture through localized inter-granular fracture.

cond-mat.mtrl-sci↗
arXiv2026

Disambiguate Gripper State in Grasp-Based Tasks: Pseudo-Tactile as Feedback Enables Pure Simulation Learning

Grasp-based manipulation tasks are fundamental to robots interacting with their environments, yet gripper state ambiguity significantly reduces the robustness of imitation learning policies for these tasks. Data-driven solutions face the challenge of high real-world data costs, while simulation data, despite its low costs, is limited by the sim-to-real gap. We identify the root cause of gripper state ambiguity as the lack of tactile feedback. To address this, we propose a novel approach employing pseudo-tactile as feedback, inspired by the idea of using a force-controlled gripper as a tactile sensor. This method enhances policy robustness without additional data collection and hardware involvement, while providing a noise-free binary gripper state observation for the policy and thus facilitating pure simulation learning to unleash the power of simulation. Experimental results across three real-world grasp-based tasks demonstrate the necessity, effectiveness, and efficiency of our approach.

cs.RO↗
arXiv2026

TDC: Sim-to-Real Transferable Directional Compliance for Contact-Rich Manipulation

Contact-rich manipulation requires robots to regulate both motion and interaction forces, yet achieving adaptive compliance remains a fundamental challenge. Learning from real-world data is costly and risky, while simulation-based approaches struggle with the sim-to-real gap in contact dynamics; existing sim-to-real methods either require real-world adaptation or sacrifice adaptive compliance by relying on isotropic compliant controllers. Our key insight is that force regulation decomposes into a time-varying but simulation-transferable directional component and a dynamics-sensitive but manually tunable magnitude component. We instantiate this directional component as two policy outputs, a task frame and a control mode vector, predicted by a visuomotor policy adapted from a pre-trained VLA model and trained via imitation learning on automatically generated simulation demonstrations. At deployment, an admittance controller integrates these predictions with human-specified stiffness and target wrench values to realize adaptive compliance. Our approach achieves adaptive compliance using only simulation data and can benefit from large-scale VLA pre-training. Extensive real-world experiments on four contact-rich tasks, microwave opening, peg-in-hole insertion, whiteboard wiping, and door opening, demonstrate strong task success rates and robustness to external disturbances. Project page: https://yifei-y.github.io/project-pages/TDC/.

cs.RO↗
arXiv2026

Rolling-WAM: World Action Models with Rolling Imagination

World Action Models (WAMs) couple action generation with future visual prediction for robotic manipulation. However, completing the joint video-action denoising process at each replanning cycle incurs substantial latency, delaying action updates and limiting closed-loop responsiveness. We present Rolling-WAM, a formulation that distributes joint denoising across successive replanning cycles. Our method maintains a sliding window of video-action chunks at staggered noise levels. At each step, a rolling noise schedule fully denoises the imminent action chunk for execution, while partially refining farther-future chunks. As the window advances with new camera observations, the retained future chunks continue their denoising process. This distributes the computational cost over time while carrying an evolving visual-action context across chunk boundaries. Evaluations on LIBERO, RoboTwin, and a real-world Unitree G1 humanoid show that Rolling-WAM achieves competitive manipulation performance. By removing the need to denoise the entire prediction horizon from scratch, it delivers a 4.5x steady-state replanning speedup over standard joint WAMs.

cs.RO↗
arXiv2026

Direct estimation and inference of differential Granger causality between two high-dimensional time series

Comparing dependence structures between two related multivariate time series is of fundamental interest in many scientific applications, where changes may occur in both directed temporal interactions and contemporaneous connectivity. Modeling each time series by a vector autoregressive (VAR) model, we propose a new framework for estimation and inference of differential Granger causality (DiffGC) and differential network (DiffNet) structures in high dimensions. The proposed method is based on a novel estimating equation derived from the Yule-Walker equations that directly links the difference between VAR transition matrices to the corresponding difference between precision matrices. Unlike separate estimation strategies that estimate the two VAR models individually and then take their difference, the proposed method directly targets the differential structures and requires sparsity only of the differences, thereby accommodating potentially dense individual networks, including hub structures. Building on the direct estimators, we develop selective inference procedures for both DiffNet and DiffGC parameters, providing valid post-selection inference while accounting for the data-driven screening process. Theoretically, we establish convergence rates and support recovery guarantees for the proposed estimators, derive asymptotic distributions for the selective inference targets, and obtain new consistency results for DiffNet estimation under weaker assumptions than existing methods. Simulation studies confirm the theoretical convergence rates and demonstrate accurate support recovery and robust inferential performance. An application to resting-state electroencephalography (EEG) data identifies substantial changes in both contemporaneous and Granger-causal connectivity across recording sessions.

stat.ME↗
arXiv2026

ECG-WM: A Physiology-Informed ECG World Model for Clinical Intervention Simulation

Electrocardiogram (ECG)-based models have achieved strong performance in diagnostic tasks, yet they remain limited in modeling how cardiac dynamics evolve under external interventions. In particular, existing approaches focus primarily on static prediction and lack mechanisms to capture ECG variations under different pharmacological conditions. In this work, we propose an ECG World Model for action-conditioned predictive simulation of cardiac electrophysiology. Moving beyond disjoint pipelines, our framework features a principled integration of physiological ordinary differential equation (ODE) priors into latent diffusion dynamics via energy regularization. This structural constraint enables the synthesis of physiologically plausible post-intervention ECG trajectories while effectively mitigating generative hallucinations. Building on this simulation process, we introduce an uncertainty-aware evaluation strategy that leverages the stochasticity of diffusion sampling to characterize both the expected clinical risk and its variability, allowing a more reliable comparative assessment of candidate interventions. We evaluate our method across diverse settings, including controlled drug-response scenarios and real-world clinical records. Beyond standard waveform metrics, experimental results demonstrate improved risk calibration and strong alignment with expert-informed treatment preferences. These results establish our approach as a robust foundation for safe and intervention-aware clinical decision support.

cs.AI↗
arXiv2026

Online Learning in Stackelberg Security Games with Time-Varying Attack Intensities

This work investigates no-regret online learning for Repeated Stackelberg Security Games (RSSGs) against attackers with time-varying attack intensities. Standard RSSG models often assume fixed attack intensities, limiting their ability to capture evolving attack capabilities. To bridge this gap, we formalize a model in which follower numbers, types, and attack intensities can vary across rounds. We derive an exact mixed-integer linear programming oracle under optimistic tie-breaking. Under full-information feedback, combining this oracle with online learning algorithm yields expected $\mathcal{O}(\sqrt{T})$ regret against adversarial attacker sequences. Under bandit feedback, where only aggregate attack is observed, we represent the payoffs of all candidate defense strategies through a small set of basis strategies. This representation enables unbiased estimation of all candidate strategies' payoffs from a single aggregate payoff per round. Building on this representation, we combine barycentric spanner with EXP2 to achieve expected $\mathcal{O}(\sqrt{T})$ regret against adversarial sequences. This guarantee contains no additional logarithmic factors in $T$ and matches an $Ω(\sqrt{T})$ lower bound established within our model. Numerical simulations demonstrate the effectiveness of the proposed algorithms.

cs.GT↗
arXiv2026

A Unified Unsupervised Framework for Genome-Wide Association Studies in Heterogeneous Populations

Genome-wide association studies (GWAS) have greatly advanced the discovery of genetic variants underlying complex traits and diseases. Yet in heterogeneous populations, existing GWAS strategies typically either pool all individuals under an assumption of population homogeneity or perform meta-analysis across predefined subgroups, both of which are limited when latent genetic heterogeneity attenuates subgroup-specific effects and masks true associations or subgroup labels are imprecise. Here we present UCALM, a unified unsupervised framework that infers genetically homogeneous subgroups directly from the data and integrates subgroup-specific GWAS with a novel layered meta-analysis method to capture both shared and subgroup-specific association signals. Through extensive simulations and analyses of large-scale human and livestock cohorts, including the UK Biobank ($n \approx 487{,}000$) and a heterogeneous pig cohort ($n \approx 85{,}000$), we demonstrate that UCALM substantially alleviated the mean genomic inflation across 24 UK Biobank traits to 1.17 compared with 1.43 for GLM and 1.37 for LDAK-KVIK, and further revealed 74 loci in the pig cohort that were previously obscured by conventional approaches. Our results establish a robust and broadly applicable strategy for association mapping in structured populations and improve the resolution of genetic signals across diverse species.

q-bio.GN↗
arXiv2026

Learning Modular Policy for Multi-Floor Object Navigation:A Factorized Framework for Diagnostic Study

Object-goal navigation (ObjectNav) in multi-floor scenarios presents a challenge due to sparse rewards caused by long-horizon decision-making. In this paper, we propose a diagnostic study based on a modular framework with an effective learnable policy to analyze failure factors in multi-floor scenarios. To achieve an effective policy for diagnosis, we design the hierarchical factorization policy that deconstructs a single global policy into an intra-floor exploration policy and an inter-floor switching policy. To providing an effective initialization for Reinforcement Learning (RL), the lightweight intra-floor policy is learned by distilling the exploration logic of Visual Language Models (VLMs). Under idealized assumptions, we show that the factorized policy is theoretically equivalent to a single global policy at the policy-representation level. Experiment results indicate that perception performance and stair climbing stability are the primary bottlenecks in multi-floor navigation.

cs.RO↗
arXiv2026

APT: Action Expert Pretraining Improves Instruction Generalization of Vision-Language-Action Policies

Vision-Language-Action (VLA) models that couple pretrained Vision-Language Models (VLMs) with continuous action experts have achieved strong manipulation performance, yet generalization to out-of-distribution (OOD) language instructions remains poor. A known challenge is the structural imbalance in VLA data, where language is far less diverse than visual and action content, making policies prone to visual shortcuts. While discrete-action methods mitigate this through vision-language co-training, continuous action experts lack such protection: they start from random initialization and learn entirely from imbalanced data, producing noisy gradients that corrupt the VLM and fail to exploit its language capability. We address this from a Bayesian perspective, factorizing the policy into a language-agnostic Vision-Action (VA) prior and a language-conditioned VLA likelihood, and propose APT, a two-stage training method emphasizing Action expert PreTraining. In Stage 1, the action expert is pretrained as a VA prior on vision-action pairs from a frozen VLM, bypassing the language imbalance. In Stage 2, language tokens are injected through a gated fusion mechanism that integrates VLM features while preserving the learned visuomotor prior. APT applies to mainstream VLA architectures, including the $π$ and GR00T-style architectures. Comprehensive experiments validate that APT achieves consistent gains on unseen instructions and compositional tasks. Project Page: https://xukechun.github.io/papers/APT/

cs.RO↗
arXiv2026

Determination of van der Waals thin film thickness via characteristic X-ray photon yield

Characteristic X ray generation is essential for producing X ray sources and for investigating light matter interactions. Here, we demonstrate that characteristic X ray generation can also be employed to determine the thickness of van der Waals thin films in a non destructive manner. Specifically, we establish a sixth order polynomial model that relates film thickness to the characteristic X ray yield. The model can both predict the X ray yield of graphite films at arbitrary thicknesses and infer the film thickness from a corresponding characteristic X ray yield. We validate the model by comparing its predicted X ray yields for graphite films with those obtained from Monte Carlo simulations. Additionally, we experimentally validated the reliability and feasibility of the proposed method for extracting graphite film thickness via X ray photon yield characterization. Our results provide a non destructive approach for determining the thickness of van der Waals thin films.

physics.app-ph↗
arXiv2026

Large Reward Models: Generalizable Online Robot Reward Generation with Vision-Language Models

Reinforcement Learning (RL) has shown strong potential for improving robotic manipulation policies, yet its practical use remains bottlenecked by the difficulty of specifying reward functions that are both semantically meaningful and reusable across tasks. In this paper, we propose Large Reward Models (LRMs), a framework that adapts foundation VLMs into frame-level reward generators for robot policy refinement. We specialize a state-of-the-art VLM on a multi-source dataset spanning real-world robot trajectories, human-object interactions, and simulated manipulation environments. Unlike prior approaches that mainly evaluate trajectories post-hoc, LRMs expose multiple reward interfaces from visual observations: progress estimation, task completion, and temporal contrastive comparison. Starting from an imitation-learned policy, we use these VLM-derived rewards to guide PPO refinement on held-out long-horizon manipulation tasks. Our experiments show that LRM progress rewards provide the strongest non-privileged online refinement signal, improving the IL baseline and narrowing the gap to privileged environment rewards. We further deploy progress rewards for progress-weighted behavioral cloning on four real-world manipulation tasks spanning two robot platforms, improving over SFT on all four tasks. These results suggest that modality-specific specialization of foundation VLMs can provide practical visual reward signals for both simulated policy refinement and physical robot self-improvement without hand-coded task rewards.

cs.RO↗