arXiv subjects

Cheng Zhang

Publications and source records attributed to Cheng Zhang.

At least 37 records · Page 2Linked to original sources

arXiv2026

Seek-and-View Reasoning for Multi-View Spatial Understanding

Existing approaches to multi-view spatial reasoning operate largely on sparse input views. Vision-language models (VLMs) are thus restricted to understand a scene and infer spatial relations within these fixed views, leading to fragile cross-view alignment and geometry-to-language bottleneck. To address these issues, we formulate a novel Seek-and-View reasoning approach to find implicit cross-view spatial evidence by locating a question-relevant view to support the spatial reasoning. To realize this approach, we propose Vantage, a training-free model-agnostic reasoning framework that pairs a VLM with a 3D foundation model: a viewpoint-grounded reasoning stage for question analysis and view planning, followed by a geometry-grounded evidence augmentation stage to effectively synthesize and incorporate visual evidence into the final reasoning. Comprehensive experiments on six VLMs demonstrate consistent improvements on five benchmarks without fine-tuning. Overall, by revealing spatial evidence through view-grounded reasoning, Vantage can largely reduce reliance on language-based cross-view alignment and improve multi-view spatial understanding. Our code is available at https://github.com/q1xiangchen/Vantage.

cs.CV↗

arXiv2026

4D-HOF: Hand-Object Flow Matching for Feed-Forward 4D Interaction Reconstruction

Existing methods for 4D hand-object reconstruction often rely on costly per-sequence optimization, while generative approaches typically synthesize interactions from random noise, which can lead to unstable interaction prediction. We introduce 4D-HOF, a feed-forward framework that reconstructs 4D hand-object interactions from coarse but informative estimates produced by vision foundation models. Concretely, we learn a conditional flow matching model that transports foundation-model-derived hand-object states toward an interaction manifold, allowing the model to correct errors in translation, rotation, and alignment in a feed-forward manner. A key advantage of our generative formulation is that it naturally enables test-time guidance within the transport process. Rather than applying a separate post-hoc optimization after reconstruction, we directly steer the evolving generative states using physical interaction constraints and observed 2D evidence, allowing the reconstruction to be refined as part of the generative process itself. By training the generative model on diverse datasets, 4D-HOF generalizes robustly to challenging in-the-wild scenarios. Experiments on out-of-domain benchmarks show that 4D-HOF achieves state-of-the-art performance, producing more stable and accurate 4D hand-object reconstructions.

cs.CV↗

arXiv2026

Feedforward Novel View Synthesis for Heterogeneous Cameras

Feed-forward novel view synthesis has recently shown promising results from sparse posed images, but most existing methods assume that context and target views share a fixed camera family. This homogeneous-camera assumption breaks in practical multi-sensor systems, where perspective, fisheye, and panoramic cameras may coexist and where the target projection may be unseen during training. We study feed-forward NVS across heterogeneous central cameras and identify a key ambiguity introduced by tokenization: a visual token aggregates a projection-dependent bundle of pixel rays, while existing camera encodings mainly expose absolute rays or token-center relations. To address this, we combine token-center relative Camera Positional Encodings and proposed local raymaps, a token-level representation that explicitly describes the intra-patch ray distribution summarized by each token. We further propose projection-aware 2D RoPE, which replaces raw image-grid coordinates with ray-induced angular coordinates so that relative positional reasoning is aligned across camera projections. Together, these components treat diverse cameras as calibrated samplings of a shared ray space rather than separate visual domains. On ScanNet++ with heterogeneous-camera system, our method improves over camera-conditioned baselines under mixed-camera evaluation and demonstrates zero-shot generalization to panoramic views.

cs.CV↗

arXiv2026

Beyond Memory: Harnessing Long-Horizon Agents with Explicit Belief States

Large language model (LLM) agents can now undertake increasingly complex tasks, but the way they organize interaction history into memory does not ensure a coherent understanding of the current world. We introduce PoS, an inference-time framework that constructs and continually maintains explicit belief states as the agent's decision context. Each belief combines an estimate of the current world state with unresolved task requirements, making explicit what the agent still needs to learn and accomplish. To keep this belief reliable and actionable, PoS validates its consistency and monitors task progress to detect Belief Trapping, where the agent continues to act without making meaningful progress toward the goal. Recovery is then tailored to both the trapping pattern and the type of unresolved task requirement. Experiments on four benchmarks spanning execution and diagnosis show that PoS achieves the highest overall performance on every benchmark with all three LLM backbones. Ablations demonstrate the importance of consistency validation and recovery, while context-scaling experiments show resilience to context growth. Together, these results support belief construction and continual maintenance as a foundation for long-horizon context management beyond history retention and compression.

cs.AI↗

arXiv2026

OmniFabric: Coherent UV Space Texture Synthesis for 3D Garment Reconstruction

Automated generation of production-ready 3D garment assets from a single image is a central challenge in digital content creation. While recent generative models have significantly advanced 3D geometry reconstruction, synthesizing high-quality textures remains a bottleneck. Existing methods often bake environmental illumination and shadows directly into the texture map, or they fail to maintain global structural coherence, making the resulting assets unusable for physical simulation and relighting. In this work, we introduce OmniFabric, a novel approach that synthesizes globally coherent texture maps directly within the 2D sewing pattern space. Given a single reference image, our pipeline utilizes an estimated 3D mesh and generative priors of powerful Vision-Language Models (VLM) to establish a complete but coarse texture initialization across the unwrapped sewing patterns. We then leverage a specialized diffusion transformer, trained via an automated synthetic data engine and conditioned on 3D positional features, to refine this initialization directly in the canonical UV domain. This effectively removes distortion and baked-in artifacts to extract a clean and normalized texture map that preserves the original garment design. Extensive experiments demonstrate that OmniFabric significantly outperforms state-of-the-art baselines, yielding photorealistic 3D garments with high-quality textures.

cs.CV↗

arXiv2026

Parameter-Efficient Construction of the Rashomon Slice for Concept Bottleneck Models

In many machine learning problems, there may exist multiple models that achieve nearly identical predictive performance while relying on fundamentally different internal logic. However, standard training procedures produce a single model, offering no practical way to explore alternatives that may better suit downstream needs. The set of these equally accurate models is known as the Rashomon set. Exploring the Rashomon set is particularly challenging in large and complex hypothesis spaces, such as Concept Bottleneck Models (CBMs), which are widely used in computer vision to make predictions through intermediate, human-understandable concepts. In this paper, we provide a method for efficiently exploring the Rashomon set of CBMs. Our framework introduces a specialized parallel parameter-efficient adaptation module, combined with a checkpointing scheme and a concept diversity objective, to generate multiple equally accurate CBMs from a single training process. Empirical results show that our method finds models with better diversity than baselines while using much less memory. We further demonstrate that access to these diverse yet accurate CBMs enables trustworthy model selection, resolution of inter-class confusion, and reliable abstention in decision-making.

cs.LG↗

arXiv2026

Endpoint eigenfunction restriction estimates in codimension two

We investigate the optimal endpoint $L^2$ restriction estimates of Laplace eigenfunctions on submanifolds of codimension 2 in a smooth closed Riemannian manifold $M$. For every fixed smooth codimension-two submanifold $Σ$, we prove a little-o improvement $o(λ^{1/2}\sqrt{\logλ})$ on the classical estimate $O(λ^{1/2}\sqrt{\logλ})$ of Burq--Gérard--Tzvetkov and Hu. Our proof uses the Bargmann transform and Tataru's phase-space representation to reduce the problem to Stein--Street's estimate for singular Radon transforms. The three-dimensional case can be handled directly by Ricci--Stein's estimate. Moreover, we construct explicit examples to show that the little-o improvement is optimal in general. These are closely related to earlier counterexamples for endpoint Strichartz estimates by Montgomery--Smith and Carbery--Hofmann. In particular, we establish the log-free estimate $O(λ^{1/2})$ when $(M,g)$ and $Σ$ are real analytic, whereas this estimate fails in general in the smooth setting.

math.AP↗

arXiv2026

Gradient Flow for the Jet Transport Coefficient: A Quenched Lattice Benchmark

We apply the gradient-flow framework to determine the nonperturbative medium contribution to the jet transport coefficient $\hat q$ using quenched lattice QCD in the temperature range $0.76T_c \leq T \leq 2.25T_c$. The calculation focuses on the leading-twist contribution at leading order for an asymptotically energetic quark propagating through a thermal pure-gluon plasma. In the infinite-energy limit, the operator-product expansion expresses $\hat q/T^3$ in pure gauge theory as the product of a perturbative short-distance coefficient and the entropy density $s/T^3$. The latter is determined from the gradient-flow energy-momentum tensor at three lattice spacings for each temperature and the former is estimated perturbatively at one-loop order. We find that $\hat q/T^3$ is strongly suppressed below $T_c$, rises rapidly across the deconfinement transition, and remains approximately constant at $\hat q/T^3\simeq1.1$ throughout the deconfined temperature range studied. This is statistically consistent with the previous quenched lattice determination based on a different renormalization procedure, providing a nontrivial validation of the gradient-flow framework for the jet transport coefficient. The systematic treatment of operator renormalization and matching provided by gradient flow becomes particularly important in full QCD, because the need to renormalize quark operators and account for their mixing with the gluonic sector can likewise be addressed within the same framework.

hep-lat↗

arXiv2026

Affine dual Minkowski problem for general measures

In 2025, the affine dual curvature measures $\widetilde{C}_m^a(K,\cdot)$ of convex body $K$ were introduced by Cai, Leng, Wu, and Xi, and the even Minkowski problem for the affine dual curvature measures was solved. In this paper, the necessary and sufficient condition of affine dual Minkowski problem for general measures with $m>1$ are proposed.

math.MG↗

arXiv2026

DITTO: Dexterous Interface for Transparent TeleOperation

Collecting data for manipulation with high-DOF hands is challenging, as interfaces must capture rich hand motion while rendering the contact interactions essential for precise manipulation. Existing data collection approaches face a trade-off: teleoperation ensures deployment consistency but lacks force feedback, while handheld (in-the-wild) systems provide natural force transparency but introduce a visual embodiment gap at deployment. We present DITTO, a Dexterous Interface for Transparent TeleOperation, which resolves this through the anatomically informed co-design of a dexterous 7-DOF robotic hand and a kinematically equivalent motorized exoskeleton. A 1-to-1 actuator mapping between the exoskeleton and robotic hand enables handheld (in-the-wild) data collection and bilateral teleoperation with joint-level force feedback unified in a single platform. We demonstrate that the DITTO exoskeleton spans the operator's natural index-to-thumb workspace, and showcase DITTO's dexterous capabilities via learned policies on contact-rich tasks.

cs.RO↗

arXiv2026

Improving precision scaling via backaction-evading continuous measurement in a driven-dissipative Kerr parametric oscillator

Dissipative phase transitions in the driven-dissipative Kerr parametric oscillator offer a promising route for realizing criticality-enhanced quantum sensing based on continuous measurements. However, achieving such enhancement through realistic measurement schemes remains an outstanding challenge. Here, we extend the backaction-evasion strategy introduced in our earlier work for the Gaussian linear case [arXiv:2511.22248 (2025)] to analyze how the quantum and classical Fisher information scale with the Kerr nonlinearity at dissipative critical points. Our results show that backaction-evading homodyne monitoring achieves enhanced photon-number scaling that surpasses the standard quantum limit, and significantly outperforms alternative protocols such as continuous photon counting. As an additional methodological contribution, we also implement and benchmark time-discrete approximation schemes with improved statistical convergence properties. We use these methods to compute the classical Fisher information for continuous homodyne detection, and demonstrate that they provide efficient access to this quantity near dissipative critical points, thereby extending the reach of existing methods.

quant-ph↗

arXiv2026

MAPS: Memory-Aware Predictive Scheduling Framework for Large Language Model Serving

The surge of large language model (LLM) applications on personal devices imposes massive, bursty workloads on cloud serving infrastructure. While prefill-decode disaggregation improves throughput and scalability, memory-bound decode instances often suffer from persistent load imbalance, as output lengths are unknown when requests arrive at the cloud. To address this, we propose MAPS, a Memory-Aware Predictive Scheduling framework tailored for disaggregated LLM serving. MAPS performs device-assisted speculative output length prediction overlapped with cloud-side prefilling, incurring negligible latency overhead. To handle generation uncertainty, MAPS applies uncertainty-aware calibration to derive output-length upper bounds with target coverage, enabling safe scheduling decisions. Building on these bounds, MAPS employs a hierarchical global-local scheduling strategy to mitigate inter-decoder queue buildup and intra-decoder head-of-line blocking. Extensive experiments on two real-world workloads and two LLMs show that MAPS significantly outperforms three state-of-the-art systems, reducing average end-to-end latency by 42.6 and tail latency by up to 84.8.

cs.AI↗

arXiv2026

Measuring Progress in Reasoning Toward Mathematical Discovery with Automatic Verification

Can AI make progress on important, unsolved mathematical problems? Large language models are now capable of sophisticated mathematical and scientific reasoning, but whether they can perform novel research is still widely debated and underexplored. We introduce HorizonMath, a benchmark of 113 predominantly unsolved problems spanning eight domains in mathematics and the mathematical sciences, paired with an open-source evaluation framework for automated verification. Our benchmark targets the generator-verifier gap: problems where discovery is hard and requires meaningful mathematical insight, but verification is computationally straightforward. This contrasts with most existing research-level benchmarks, which instead rely on formal proof verification or manual review, both of which are expensive to scale. Because these solutions are unknown, HorizonMath is resistant to data contamination, and most state-of-the-art models score under 10%. Using this framework, we identify six novel solutions to research problems that either resolve previously open questions or improve on the best-known published results, with GPT-5.4 Pro and GPT-5.6 Sol each discovering three of these solutions. Across seven frontier model families, reasoning efficiency and behavior also vary substantially. We release HorizonMath as an open challenge and a growing community resource, where each verified solution is a candidate contribution to the mathematical literature.

cs.LG↗

arXiv2026

Synergising Local Geo-Environmental Characteristics with Spatial Context for Enhancing Landslide Susceptibility Mapping

Data-driven methods are widely used in landslide susceptibility mapping (LSM) because they can effectively model the complex relationships between landslides and geo-environmental conditions. Existing data-driven approaches generally follow two types of data representations. Pixel-based models focus solely on the geo-environmental characteristics of a specific landslide but neglect the influence of its surrounding environment. Patch-based models incorporate surrounding spatial context but may include pixels with weak or no spatial relevance to the target landslide location. To address this limitation, this study proposes a Local-Geo and Spatial Context Fusion (LGSCF) strategy, which synergises the geo-environmental characteristics of landslide points with their corresponding spatial context through a feature-wise modulation mechanism. We tested the LGSCF strategy by integrating it into several representative convolutional neural network (CNN) architectures, creating nine different LGSCF-based models. The primary study area covers approximately 2644 km2 across Jenai and Sinyi Townships in Nantou County, Taiwan, and the dataset comprises 5332 landslide samples and an equal number of non-landslide samples. The results show that LGSCF-based models consistently outperform their corresponding baselines, achieving F1-scores up to 87.09% and AUC values up to 0.9472. Furthermore, the susceptibility maps produced by LGSCF-based models show that known landslides are more accurately concentrated in "very high" susceptibility zones with fewer misclassifications. These findings demonstrate that our fusion strategy can significantly improve the accuracy of landslide susceptibility mapping.

cs.CV↗

arXiv2026

Eigenvalues and eigenfunctions of the fractional Laplacian on the interval

We prove a three-term asymptotic formula for the eigenvalues of the fractional Laplacian on the bounded interval $(-1,1)$. This improves the eigenvalue asymptotics of Kulczycki--Kwaśnicki--Małecki--Stós and Kwaśnicki, and confirms the conjectural $O_α(n^{-2})$ remainder suggested by the numerical simulations of Kaleta--Kwaśnicki--Małecki. Moreover, we prove that the normalized eigenfunctions are bounded uniformly in the eigenvalue index $n$ and the fractional order $α$. This settles the conjecture proposed by Kwaśnicki through numerical experiments. Furthermore, we prove that the $n$-th eigenfunction has exactly $n-1$ zeros in the interval $(-1,1)$ and every zero is simple, and hence there are exactly $n$ nodal domains. A key ingredient in the proof is an explicit representation of the eigenfunction.

math.CA↗

arXiv2026

The conservative turn in science: The changing character of knowledge recombination

This study examines how the dominant mode of cross-disciplinary knowledge combination has changed over time. It addresses a limitation of existing studies that document declining disruptiveness at the level of individual works but do not reveal whether the mechanisms of cross-disciplinary knowledge flow have themselves shifted. Drawing on 56 million publications and 816 million citations from OpenAlex (1960-2025), the study classifies cross-disciplinary citations into five types based on the set relationship between the subfield portfolios of citing and cited publications and introduces a semantic-distance measure derived from SPECTER-v2 embeddings of publication abstracts. It subsequently demonstrates that cross-disciplinary citations have shifted markedly away from configurations with no shared disciplinary ground toward types built on partially overlapping knowledge bases, a reorientation further reflected in a steady decline in average semantic distance between citing and cited fields. This shift co-occurs with a diffusion of brokerage positions across the network and a greater reliance on older literature, consistent with a system that has become broader and more decentralized in its interdisciplinary reach. Finally, this study discusses the implications of these findings for science policy and for the design of cross-disciplinary funding mechanisms.

cs.SI↗

arXiv2026

X2-N: A Transformable Wheel-legged Humanoid Robot with Dual-mode Locomotion and Manipulation

Wheel-legged robots combine the efficiency of wheeled locomotion with the versatility of legged systems, enabling rapid traversal over both continuous and discrete terrains. However, conventional designs typically employ fixed wheels as feet and limited degrees of freedom (DoFs) at the hips, resulting in reduced stability and mobility during legged locomotion compared to humanoids with flat feet. In addition, most existing platforms lack a full upper body with arms, which limits their ability to perform dexterous manipulation tasks. In this letter, we present X2-N, a high-DoF transformable robot with dual-mode locomotion and manipulation. X2-N can operate in both humanoid and wheel-legged forms and transform seamlessly between them through joint reconfiguration. We further propose a reinforcement learning (RL)-based whole-body control framework tailored to this morphology, enabling control across hybrid locomotion, transformation, and manipulation. We validate X2-N in a range of challenging locomotion and manipulation tasks, including dynamic skating-like motion, stair climbing, and package delivery. Results demonstrate high

cs.RO↗

arXiv2026

Lattice KP type equations arising from eigenfunctions and Dbar problem

In this paper, we construct the lattice Kadomtsev-Petviashvili (KP) type eigenfunction equations. A homogeneous nonlocal $\bar{\partial}$ problem is considered, from which we are able to define the eigenfunction of the Lax pair of the lattice KP equation. The eigenfunction together with its expansions at infinity and at a finite analytic point provide formulations of the lattice modified KP equation, the lattice Schwarzian KP equation and the Nijhoff-Quispel-Capel KP (NQC-KP) equation. We also consider an inhomogeneous nonlocal $\bar{\partial}$ problem. It defines the eigenfunction of the Lax pair of the lattice modified KP equation. This eigenfunction generates a direct formulation for the NQC-KP equation, which is different from the previous ones. Explicit solutions of these equations are obtained, from which we can see the difference of the different formulations for same equations.

nlin.SI↗