arXiv subjects

Mingxin Wang

Publications and source records attributed to Mingxin Wang.

At least 19 recordsLinked to original sources

arXiv2026

SoftVTBench: A Deformation-Aware Visuo-Tactile Dataset and Benchmark for Deformable-Object Manipulation

Physical interaction quality is central to deformable-object manipulation, yet most benchmarks evaluate task success alone. A policy may complete the task while allowing slip or causing excessive compression. A primary bottleneck is the absence of visuo-tactile datasets that pair policy-visible contact observations with independent physical ground truth over complete tasks. We introduce SoftVTBench, a visuo-tactile dataset for physical-interaction-aware deformable-object manipulation. It contains 4,000 expert demonstrations and more than 50 assets, including volumetric deformable objects and visually matched rigid twins. At 20 Hz, each episode synchronizes multi-view RGB, dual-finger tactile RGB and marker motion, proprioception, language, and binary and continuous gripper actions, alongside evaluator-only finite-element (FEM) states. Building upon this dataset, we establish a closed-loop benchmark that uses fixed object-specific calibration to define the Deformation-aware Success Rate (DSR), which counts a rollout as successful only when it completes the task and keeps peak normalized deformation within tolerance. Across Diffusion Policy, $\pi_{0.5}$, and FastWAM, all 12 in-distribution configurations contain successful rollouts that violate the deformation tolerance, accounting for 0.7--24% of each configuration's successes. Under distribution shift, visuo-tactile variants achieve higher task success in all six policy--suite comparisons and higher DSR in five, whereas their in-distribution benefits are mixed. These results show that making touch available does not by itself ensure effective multimodal fusion. SoftVTBench therefore provides a common visuo-tactile resource for studying not only whether a policy succeeds, but how it physically interacts with deformable objects and when touch improves that interaction.

cs.RO↗

arXiv2026

Diffusion contrast induced dynamics in a time-periodic competition-diffusion system with equal total resources

This paper investigates a time-periodic, spatiotemporally heterogeneous competition-diffusion system, where two species have different intrinsic growth rates but equal total resources per period. Using asymptotic analysis and principal eigenvalue theory, we systematically examine its dynamics. When both diffusion rates are small, spatial heterogeneity of the difference in time-averaged resources leads to uniform persistence, stable coexistence, and asymptotic spatial segregation; whereas its spatial uniformity produces multiple regimes--fast-diffuser selection, parameter-dependent coexistence, or slow-diffuser dominance--separated by smooth threshold curves under different structural conditions. In the regime where at least one diffusion rate is large, we obtain a detailed classification of the local dynamics, with sharp delineation of parameter regions for stable coexistence, competitive exclusion, and a novel bistable scenario, where both semi-trivial periodic solutions are linearly stable while an unstable coexistence state also exists. We further characterize the asymptotic profiles of positive solutions in the mixed-scale diffusion limit. Our results show that the interplay between temporal periodicity and spatial heterogeneity can overturn the classical "slower diffuser always prevails" principle, leading to a substantially richer range of ecological outcomes.

math.AP↗

arXiv2026

Push-Wiper: Toward General-Purpose Robotic Cleaning across Varied Stains and Surfaces with Segmented Pushing Trajectories

Viscous stains, characterized by high viscosity and complex rheological properties, remain a major challenge for robotic surface cleaning. Conventional wiping often spreads the stain, while scrubbing provides stronger friction but risks damaging the surface. In this paper, we propose Push-Wiper, a framework that reformulates viscous stain cleaning as an aggregation problem. Push-Wiper employs a sponge to progressively gather stains through segmented pushing trajectories, followed by a post-processing phase that detaches the aggregated material and enables sponge self-cleaning. We adopt a stepwise strategy for stain gathering and leverage Diffusion Policy to generate adaptive pushing action sequences. These sequences are executed through our Arbitrary Surface Pose Interpolator (ASPI) and a hybrid force-position controller, allowing the method to generalize to stains with diverse spatial distributions. Push-Wiper achieves a cleaning score (CS), defined as the percentage of stain area removed, up to 130% higher than baseline methods. Without additional training, Push-Wiper also transfers in a zero-shot manner to solid residues, liquid spills, unseen viscous stains, and curved surfaces with varying geometries. Our experiments demonstrate the cleaning effectiveness of Push-Wiper and its strong generalization ability. The project website is available at https://push-wiper.github.io/.

cs.RO↗

arXiv2026

ST-WAM: Semantic-Temporal World Action Model for Robust Manipulation under Visual Distribution Shifts

World Action Models (WAMs) have emerged as a promising paradigm by jointly modeling robot actions and future visual dynamics. However, their reliance on pixel-generative future supervision can entangle action-relevant state transitions with task-irrelevant visual content, limiting robustness under visual distribution shifts. We identify Training-Distribution Hallucination, a recurring phenomenon in which futures conditioned on visually shifted observations hallucinate training-domain content rather than remain faithful to the current scene. A controlled frame-triplet diagnosis further shows that DINOv3 features remain more stable across visual shifts while better preserving task-state distinctions than Wan-VAE latents. Rather than correcting the predicted futures, we propose Semantic-Temporal WAM (ST-WAM) to improve action robustness by using DINOv3 as a shared semantic representation for future prediction and history retrieval while retaining fine-grained VAE dynamics. Its Dual-Space Future Experts (DSFE) jointly predict future VAE latents and DINO features, while Current-Anchored Intent Retrieval (CAIR) retrieves task-relevant evidence from recent DINO history under the current visual-language context. ST-WAM is trained end-to-end without additional embodied pretraining or task-specific annotations, and requires no explicit future generation at inference. It achieves 98.7% on LIBERO and 92.8% on RoboTwin 2.0; more importantly, compared with Fast-WAM, it improves zero-shot LIBERO-Plus performance by 21.3 percentage points and more than doubles real-world success under visual shifts from 25.8% to 61.5%. These results demonstrate that semantic-temporal modeling effectively complements pixel-generative dynamics for robust manipulation.

cs.RO↗

arXiv2026

MASTE: A Multi-Agent Pipeline for Zero-Shot Aspect Sentiment Triplet Extraction

Aspect Sentiment Triplet Extraction (ASTE) requires jointly identifying (aspect, opinion, sentiment) triples from a given review sentence. While large language models (LLMs) achieve strong zero-shot performance on many NLP benchmarks, their effectiveness on ASTE remains limited, as single-pass generation forces the model to determine span boundaries, opinion grouping, and sentiment polarity in a single decoding step. Common remedies, such as few-shot in-context learning and chain-of-thought prompting, offer only marginal improvements and rely heavily on either in-domain demonstrations sampled from labeled training data or carefully engineered reasoning prompts, neither of which is broadly available in zero-shot deployment. Inspired by the classical agent paradigm, we propose MASTE, a multi-agent pipeline for zero-shot Aspect Sentiment Triplet Extraction. MASTE decomposes ASTE into four sequential stages, where specialized agents handle different compositional subtasks with explicit conditioning on prior outputs. This design enables entirely training-free zero-shot ASTE and generalizes across different backbones and datasets. Extensive experiments on four ASTE benchmarks show that MASTE substantially outperforms zero-shot and chain-of-thought LLM baselines under the same backbone, narrowing the gap to fully supervised methods without using any labeled triplets. Code is available at https://github.com/Hankerlove/MASTE.

cs.CL↗

arXiv2026

SoftVTBench: A Safety-Aware Visuo-Tactile Benchmark for Physically Constrained Robotic Manipulation of Deformable Objects (Early Version)

Deformable object manipulation poses challenges beyond task completion: successful execution must also maintain safe physical interaction, holding the object stably without slip or drop while avoiding excessive deformation. However, existing manipulation benchmarks are predominantly success-oriented and rarely evaluate whether a policy remains physically safe throughout execution. We present SoftVTBench, a safety-aware visuo-tactile benchmark for physically constrained deformable object manipulation. Built in Isaac Sim with finite-element-simulated deformable objects, SoftVTBench provides multi-view RGB observations, RGB tactile sensing with marker motion, proprioception, and language instructions, and defines four matched task suites over object type (deformable vs. rigid) and variation axis (object vs. spatial). It separately reports Goal Success and Safety Success; the latter additionally requires no drop and peak deformation below a calibrated object-specific threshold, measured from policy-hidden privileged Finite Element Method (FEM) states. We implement pi0.5-based baselines under this protocol. Experiments show that success-only evaluation substantially overstates policy performance, as a large fraction of goal-completing rollouts still violate physical safety. Furthermore, incorporating tactile sensing improves Safety Success (e.g., from 21.4% to 35.6% on object-centric deformable tasks) and reduces object deformation during execution, while maintaining comparable Goal Success. SoftVTBench provides a reproducible benchmark for studying visuo-tactile deformable manipulation under physical interaction constraints.

cs.RO↗

arXiv2026

ABot-M0.5: Unified Mobility-and-Manipulation World Action Model

Mobile manipulation is a key capability for general-purpose robots, yet remains challenging for current embodied learning methods. VLA policies are typically reactive and lack explicit world modeling, while existing World Action Models (WAMs) are still poorly aligned with the structure of mobile manipulation: they operate on coarse video chunks, model entangled navigation-manipulation actions, and train inverse dynamics under supervision that does not match autoregressive inference. As a result, they often miss fine-grained contact dynamics, suffer from action-distribution conflicts, and accumulate errors over long-horizon rollouts. We propose ABot-M0.5, a new WAM built on the insight that mobile manipulation requires alignment at three levels: temporal granularity, action space, and train-test consistency. To align temporal granularity, we introduce intermediate latent actions that capture local visual state transitions and serve as an bridging action space between video latents and embodiment-specific controls. To align action space, we design a dual-level Mixture-of-Transformers architecture that disentangles both modality representations and heterogeneous action subspaces such as base movement and arm manipulation. To align inference conditions, we propose the dream-forcing training strategy that progressively trains inverse dynamics on model-predicted videos, improving train-test alignment and robustness during autoregressive prediction. Experiments on challenging mobile and fine-grained manipulation benchmarks demonstrate that ABot-M0.5 achieves state-of-the-art performance in both long-horizon task success and finegrained control accuracy. These results highlight the critical importance of granularity-aligned, action-disentangled, and inference-consistent world-action modeling.

cs.CV↗

arXiv2026

AtomEval: Validity-Aware Atomic Evaluation of Adversarial Claim Rewriting in Fact Verification

Large language models (LLMs) can rewrite refuted claims to evade evidence-based fact verifiers, but conventional attack success rate (ASR) can be inflated when rewrites change, weaken, or correct the false proposition they are supposed to preserve. We introduce AtomEval, a validity-aware evaluation protocol for fixed-evidence adversarial claim rewriting. AtomEval represents claims as subject--relation--object--modifier (SROM) atoms, applies a one-way preservation gate to separate valid verifier evasion from proposition-changing rewrites, and reports validity-aware attack success rate (VASR), which counts only verifier-evasive rewrites that preserve the original false proposition. AtomEval further provides fine-grained diagnostics that explain both proposition-level failures and non-minimal valid rewrites. On FEVER refuted-claim rewriting, AtomEval exposes and explains ASR inflation: many apparent attacks fool the verifier by altering, weakening, or correcting the proposition they should preserve. By making attacked-proposition preservation explicit and measurable, AtomEval provides a stable evaluation target for evaluating adversarial rewriters that must balance verifier evasion with proposition preservation.

cs.CL↗

arXiv2026

PhaForce: Phase-Scheduled Visual-Force Policy Learning with Slow Planning and Fast Correction for Contact-Rich Manipulation

Contact-rich manipulation requires not only vision-dominant task semantics but also closed-loop reactions to force/torque (F/T) transients. Yet, generative visuomotor policies are typically constrained to low-frequency updates due to inference latency and action chunking, underutilizing F/T for control-rate feedback. Furthermore, existing force-aware methods often inject force continuously and indiscriminately, lacking an explicit mechanism to schedule when / how much / where to apply force across different task phases. We propose PhaForce, a phase-scheduled visual--force policy that coordinates low-rate chunk-level planning and high-rate residual correction via a unified contact/phase schedule. PhaForce comprises (i) a contact-aware phase predictor (CAP) that estimates contact probability and phase belief, (ii) a Slow diffusion planner that performs dual-gated visual--force fusion with orthogonal residual injection to preserve vision semantics while conditioning on force, and (iii) a Fast corrector that applies control-rate phase-routed residuals in interpretable corrective subspaces for within-chunk micro-adjustments. Across multiple real-robot contact-rich tasks, PhaForce achieves an average success rate of 86% (+40 pp over baselines), while also substantially improving contact quality by regulating interaction forces and exhibiting robust adaptability to OOD geometric shifts.

cs.RO↗

arXiv2025

Stock Prediction via a Dual Relation Fusion Network incorporating Static and Dynamic Relations

Accurate modeling of inter-stock relationships is critical for stock price forecasting. However, existing methods predominantly focus on single-state relationships, neglecting the essential complementarity between dynamic and static inter-stock relations. To solve this problem, we propose a Dual Relation Fusion Network (DRFN) to capture the long-term relative stability of stock relation structures while retaining the flexibility to respond to sudden market shifts. Our approach features a novel relative static relation component that models time-varying long-term patterns and incorporates overnight informational influences. We capture dynamic inter-stock relationships through distance-aware mechanisms, while evolving long-term structures via recurrent fusion of dynamic relations from the prior day with the pre-defined static relations. Experiments demonstrate that our method significantly outperforms the baselines across different markets, with high sensitivity to the co-movement of relational strength and stock price.

cs.LG↗

arXiv2025

Symmetry-breaking bifurcation of periodic solutions for a free-boundary tumor model

In this paper, we consider a free boundary multi-layer tumor model that incorporates a $T-$periodic provision of external nutrients $\Phi(t)$. The simplified model contains three parameters: the mean of periodic external nutrients $\Phi(t)$, the threshold concentration $\widetilde{\sigma}$ for proliferation and the cell to cell adhesiveness coefficient $\gamma$. We first study the flat solution and give a complete classification about $\frac{1}{T} \int_0^T \Phi(t) d t$ and $\widetilde{\sigma}$ according to global stability of zero equilibrium solution or global stability of the positive periodic solution. Precisely, (i) a zero flat solution is globally stable under the flat perturbations if and only if $\widetilde{\sigma} \geqslant \frac{1}{T} \int_0^T \Phi(t) d t$; (ii) If $\widetilde{\sigma}<\frac{1}{T} \int_0^T \Phi(t) d t$, then there exists a unique positive flat solution $\left(\sigma_*(y, t), p_*(y, t), { \rho_*(t)}\right)$ with period $T$ and it is a global attractor of all positive flat solutions for all $\gamma>0$. We further investigate periodic solutions bifurcating from the flat periodic solution $\left(\sigma_*(y, t), p_*(y, t), { \rho_*(t)}\right)$. By periodicity and symmetry, we not only give symmetry-breaking periodic solutions for all positive parameter $\gamma_j$, but also show the existence of a plethora of periodic bifurcations. For the free boundary tumor problem, this is the first result of the existence of periodic bifurcations.

math.AP↗

arXiv2025

Global boundedness and finite time blow-up of solutions for a quasilinear chemotaxis-May-Nowak model

In this paper, we introduce the nonlinear diffusion term $\nabla\cdot(D(u)\nabla u)$ into the chemotaxis-May-Nowak model to investigate the effects of $D(u)$ and chemotaxis on the global existence, boundedness, and finite time blow-up of solutions. Here, $D(u)$ generalizes the prototype $(1+u)^{m-1}$ with $m\in\R$. For the parabolic-elliptic-parabolic case, if $m>2+\frac{n}{2}-\frac2{n}$ when $n\ge3$ and $m>\frac32$ when $n=2$, then all solutions exist globally and remain bounded, whereas if $n\in\{2,3\}$ and $m<1$, finite time blow-up occurs when $\Omega$ is a ball and the initial data are radially symmetric. For the fully parabolic case, if $m>1+\frac{n}{2}-\frac2{n}$, then all solutions exist globally and remain bounded.

math.AP↗

arXiv2025

Generalized principal eigenvalue of time-periodic cooperative nonlocal dispersal systems and applications

It is well known that, in the study of the dynamical properties of nonlinear reaction-diffusion systems, the sign of the principal eigenvalue of the linearized system plays an important role. However, for the nonlocal dispersal systems, due to the lack of compactness, the essential spectrum appear, and the principal eigenvalue may not exist. In this paper, by constructing monotonic upper and lower control systems, we obtain the generalized principal eigenvalue of the cooperative irreducible system and demonstrate that this generalized principal eigenvalue plays the same role as the usual principal eigenvalue.

math.AP↗

arXiv2025

Approximation of the generalized principal eigenvalue of cooperative nonlocal dispersal systems and applications

It is well known that, in the study of the dynamical properties of nonlinear evolution system with nonlocal dispersals, the principal eigenvalue of linearized system play an important role. However, due to lack of compactness, in order to obtain the existence of principal eigenvalue, certain additional conditions must be attached to the coefficients. In this paper, we approximate the generalized principal eigenvalue of nonlocal dispersal cooperative and irreducible system, which admits the Collatz-Wielandt characterization, by constructing the monotonic upper and lower control systems with principal eigenvalues; and show that the generalized principal eigenvalue plays the same role as the usual principal eigenvalue.

math.CA↗

arXiv2025

On a reaction-diffusion virus model with general boundary conditions in heterogeneous environments

To describe the propagation of West Nile virus and/or Zika virus, in this paper, we propose and study a time-periodic reaction-diffusion model with general boundary conditions in heterogeneous environments and with four unknowns: susceptible host, infectious host, susceptible vector and infectious vector. We can prove that such problem has a positive time periodic solution if and only if host and vector persist and the basic reproduction ratio is greater than one, and moreover the positive time periodic solution is unique and globally asymptotically stable when it exists.

math.AP↗

arXiv2025

Longtime behaviors of a reducible cooperative system with nonlocal diffusions and free boundaries

This paper aims at understanding the longtime behaviors of a reducible cooperative system with nonlocal diffusions and different free boundaries, describing the interactions of two mutually beneficial species. Compared with the irreducible and monostable cooperative system, the system we care about here has many nonnegative steady states, leading to much different and rich longtime behaviors. Moreover, since the possible nonnegative steady states on half space are non-constant, we need to employ more detailed analysis to understand the corresponding steady state problems which in turn helps us to derive a complete classification for the longtime behaviors of our problem. The spreading speeds of free boundaries and more accurate limits of $(u,v)$ as $t\to\infty$ are also discussed, and accelerated spreading can happen if some threshold conditions are violated by kernel functions.

math.AP↗

arXiv2024

A Novel Low-Background Photomultiplier Tube Developed for Xenon Based Detectors

Photomultiplier tubes (PMTs) are essential in xenon detectors like PandaX, LZ, and XENON experiments for dark matter searches and neutrino properties measurement. To minimize PMT-induced backgrounds, stringent requirements on PMT radioactivity are crucial. A novel 2-inch low-background R12699 PMT has been developed through a collaboration between the PandaX team and Hamamatsu Photonics K.K. corporation. Radioactivity measurements conducted with a high-purity germanium detector show levels of approximately 0.08 mBq/PMT for $\rm^{60}Co$ and 0.06~mBq/PMT for the $\rm^{238}U$ late chain, achieving a 15-fold reduction compared to R11410 PMT used in PandaX-4T. The radon emanation rate is below 3.2 $\rm \mu$Bq/PMT (@90\% confidence level), while the surface $\rm^{210}Po$ activity is less than 18.4 $\mu$Bq/cm$^2$. The electrical performance of these PMTs at cryogenic temperature was evaluated. With an optimized readout base, the gain was enhanced by 30\%, achieving an average gain of $4.23 \times 10^6$ at -1000~V and -100~$^{\circ}$C. The dark count rate averaged 2.5~Hz per channel. Compactness, low radioactivity, and robust electrical performance in the cryogenic temperature make the R12699 PMT ideal for next-generation liquid xenon detectors and other rare event searches.

physics.ins-det↗

arXiv2024

Longtime behaviors of an epidemic model with nonlocal diffusions and a free boundary: rate of accelerated spreading

This is the third part of our series of work devoted to the dynamics of an epidemic model with nonlocal diffusions and free boundary. This part is concerned with the rate of accelerated spreading for three types of kernel functions when spreading happens. By constructing the suitable upper and lower solutions, we get the rate of the accelerated spreading of free boundary, which is closely related to the behavior of kernel functions near infinity. Our results indicate that the heavier the tail of the kernel functions are, the faster the rate of accelerated spreading is. Moreover, more accurate spreading profiles for solution component $(u,v)$ are also obtained.

math.AP↗