Searcharxiv⌕ Search

arXiv subjects

Xudong Zhang

Publications and source records attributed to Xudong Zhang.

At least 19 recordsLinked to original sources
arXiv2026

Bridging Low-Level Image Fusion and High-Level Semantic Understanding

Remote sensing image fusion and semantic interpretation are usually optimized separately, so reconstruction quality need not preserve downstream retrieval utility. We propose LANCE, which uses structured language as an explicit, indexable semantic state. A frozen vision-language model maps the input to a closed-vocabulary text record; a language-conditioned generator fuses the imagery; and a dense verifier with a differentiable continuous summarizer re-describes the output. Text-text consistency and anti-collapse contrastive text-mask losses with negative queries close the loop without dense annotations or VLM fine-tuning. A variational EM formulation yields a per-step contraction result under regularity assumptions, distinguishing semantic faithfulness from verifier overfitting. Across seven test sets on GF2, QuickBird, WorldView-III, and WorldView-II, LANCE improves reduced-scale reference metrics by up to 11.7% (median 2.1%) over the strongest learning baseline. Independent evaluation using a held-out VLM and a supervised aerial detector yields gains of +6.4 mAP and +0.072 open-vocabulary F1 over the closest competitor. Text perturbations, cross-verifier transfer, and pseudo-RGB ablations examine the mechanisms underlying these gains. The framework connects image fusion to verifiable semantic representations for multimodal retrieval and knowledge management.

cs.AI↗
arXiv2026

IR275K: A Benchmark for Infrared Multi-Frame Super-Resolution Toward Efficient Remote Sensing

Infrared multi-frame super-resolution (MFSR) exploits complementary observations to recover spatial detail, but its evaluation remains fragmented across datasets and protocols. A common benchmark is needed to evaluate reconstruction under weak texture, varying thermal contrast, and noise in the source imagery. It should also characterize reconstruction performance across different inter-frame displacement ranges. We introduce IR275K, a curated benchmark containing 594 in- frared video sequences and 275,196 frames. It provides sequence- level train/validation/test splits and a reproducible x4 evaluation protocol. Evaluation spans single-frame and multi-frame recon- struction using seven learned baselines, the CGMamba reference model, and bicubic interpolation. All learned methods are evalu- ated under both bicubic degradation and synthetic mixed sensor noise.EDVRleadstheoverallrankingunderbicubicdegradation, whereas MSDTGP leads under mixed sensor noise. DBSR leads thehigh-displacementUAVsubsetunderbothsettings.Controlled ablations using CGMamba show that real neighboring frames improve PSNR over center-frame repetition by 1.04 dB under bicubic degradation and 0.95 dB under mixed sensor noise. Center-frame conditioning also improves reconstruction under both settings. The benchmark thus reveals how reconstruction rankings and the benefits of neighboring frames depend on evaluation conditions. The complete IR275K dataset is publicly available at: https://github.com/InfraRecon7/IR275K.

cs.CV↗
arXiv2026

Motion Retargeting with Spatiotemporal Preferences for Humanoid Loco-Manipulation

Human motion retargeting bridges the embodiment gap between humans and humanoid robots by converting human demonstrations into reference trajectories for whole-body control and loco-manipulation. For object-interaction tasks, successful execution depends on accurately preserving body--object relations at critical stages. However, uniform per-frame iteration budgets do not account for these localized accuracy requirements, limiting the computational efficiency of retargeting. To address this limitation, we present Spatiotemporal Retargeting (STR), which uses keyframes and key body parts to allocate optimization effort. Specifically, a vision-language model selects relevant motion signals and body parts, and keyframes are identified from changes in the selected signals. Based on these selections, STR assigns additional solver iterations to keyframes and higher interaction weights to the selected parts. To further improve contact fidelity, we introduce Mechanics-Aware Contact Anchoring (MCA) to refine object-surface contact positions using an approximate object-wrench model while retaining the selected surface directions. Building on the resulting references, we develop Spatiotemporal Reinforcement Learning (STRL), which combines adaptive key-stage sampling with a contact-position reward based on target proximity for downstream control. Simulation experiments on a Unitree G1 and real-world experiments on a Unitree R1 demonstrate that STR improves geometric contact coverage and continuity at lower computational cost, while the resulting references improve downstream task completion across diverse object-interaction motions.

cs.RO↗
arXiv2026

RBGA: Residual-Budget Guide Arbitration for Offline-to-Online Safe Reinforcement Learning in Dexterous Manipulation

Dexterous manipulation requires robots to leverage existing datasets for rapid skill acquisition while maintaining safety during real-world deployment to prevent damage to the robot or its surroundings. Offline-to-online safe reinforcement learning (O2O safe RL) provides a promising solution for this problem by leveraging prior experience for online improvement under safety constraints, thereby mitigating the sample inefficiency and exploration risk of learning safe policies from scratch. However, existing O2O safe RL approaches do not explicitly account for the safety budget already consumed when making decisions within an episode. Excessive cost incurred early can therefore leave little room for subsequent exploration and policy improvement. To address this limitation, we propose \textbf{R}esidual-\textbf{B}udget \textbf{G}uide \textbf{A}rbitration (RBGA), an algorithm that explicitly tracks the remaining episode safety budget and uses it to adaptively select between offline guidance and the evolving online policy. At each step, RBGA compares the cost-to-go estimates of the guide and online actions against the remaining budget, enabling step-wise action arbitration throughout the episode. We provide theoretical analysis characterizing the safety implications of residual-budget-based selection and its effect on guide-assisted exploration. Extensive experiments demonstrate that RBGA enables effective online policy improvement while maintaining strong safety performance. Our code is available at this \href{https://github.com/sii-pilab}{link}.

cs.LG↗
arXiv2026

Beyond Fixed Acquisition Functions: Temporal Difference Active Learning for Perturbation Discovery

Genetic and drug perturbation screens are important tools for elucidating disease mechanisms and validating therapeutic targets, but experimental costs limit the number of candidates that can be tested. Existing perturbation screening methods typically update the surrogate model with feedback while keeping the acquisition function fixed, limiting their ability to adapt selection to different response landscapes. Experimental feedback reveals current hits and informs subsequent selection by updating candidate predictions, providing a training signal for learning how to select experiments. We formulate budgeted perturbation discovery as a sequential decision problem and propose Temporal Difference Active Learning (TDAL), which learns an acquisition function online through temporal difference learning from a single screening trajectory, accounting for current hits and future discoveries. TDAL uses the Hellinger distance between hit probabilities for unmeasured candidates before and after an experiment to weight successor value, and reuses measured outcomes through leave-batch-out (LBO) to construct virtual transitions and alleviate training data scarcity. Across eight public screens spanning single, double, and triple perturbations under varied experimental configurations, TDAL achieves an average rank of 1.63 among eleven methods and ranks first on four tasks, demonstrating its adaptability across response landscapes.

cs.LG↗
arXiv2026

LWMa: Distributed Deployment of an Angle-Delay-Domain Wireless Foundation Model for Multiple Tasks

The growing diversity of wireless services in sixth-generation (6G) wireless communication systems calls for reusable channel representations that support multiple tasks across different propagation environments. However, deploying such representations between user equipment and the base station in frequency-division duplexing systems faces uplink feedback constraints. This paper proposes a distributed framework based on an angle-delay-domain large wireless model (LWMa) for multiple tasks. LWMa exploits angle-delay structure and delay truncation to improve feature extraction and reduce computational cost. A scenario-specific network inspired by Perceiver IO (input-output) compresses embeddings for channel reconstruction tasks, while other tasks use a low-dimensional compact global embedding. Embedding reconstruction training followed by channel-quality-guided fine-tuning preserves information useful for downstream processing. Simulations demonstrate that the proposed method supports effective channel reconstruction and multiple tasks under limited feedback with reduced deployment complexity. Further evaluations show that the recovered embeddings generalize to additional tasks and support split inference.

eess.SP↗
arXiv2026

Distance Structure of Insertion Kernels and Local Port Response in Non-Reciprocal Su--Schrieffer--Heeger Chains

Non-reciprocal lattices offer a flexible setting for directional wave transport, raising the question of how enhanced internal scattering translates into transmission between fixed external ports. We address this question for non-reciprocal Su--Schrieffer--Heeger (SSH) chains with staggered gain and loss by relating the reference-dependent insertion kernel $K_p=VG_pV$ to the coherent transmission intensity $T$. On a fixed finite graph, a strong-insertion expansion gives a contribution proportional to $γ^{1-d}$, whose coefficient is the signed sum of shortest directed paths of length $d$. When this sum is nonzero, it sets the leading behavior independently of fixed diagonal losses; when it cancels, those losses can enter the first surviving higher-order term. For a fixed cross-sublattice endpoint channel, transmission has a strictly negative slope at every nonzero differentiable stationary point of the kernel magnitude. In the weak-insertion regime, the reference-system coefficient $χ=-\mathrm{Re}\,\mathrm{Tr}[(G_{0p}Z)^2]$ predicts the initial transmission change for symmetric SSH ports, and $χ>0$ together with a strictly stable baseline guarantees local stable enhancement. Finite-chain calculations and driven dynamics connect these results to observable response and distinguish asymptotic convergence from transient settling. A strictly dissipative short-chain example exhibits local enhancement without net power gain. Together, these results separate graph-path structure, coherent port response, and dynamical establishment, providing channel-specific criteria for interpreting internal scattering.

quant-ph↗
arXiv2026

Pistis Technical Report

We introduce the Pistis model family, comprising 27B- and 9B-parameter multimodal large language models built on Qwen3.6 and Qwen3.5, respectively, and developed through a general and scalable post-training framework. The framework first establishes a strong foundation through large-scale multimodal supervised fine-tuning (SFT). Building on this SFT foundation, we propose Interleaved Distillation and Reinforcement Learning (IDRL), a novel post-training paradigm that tightly integrates on-policy distillation and reinforcement learning within a single training loop. By alternating between the two objectives, rather than optimizing either in isolation or combining them in a static joint loss, IDRL enables more effective knowledge transfer, greater optimization stability, and more precise credit assignment for long-horizon agentic trajectories, leading to stronger performance while mitigating common capability trade-offs. At both model scales, the framework produces two specialized variants: Pistis-Thinking, designed to strengthen deep multimodal reasoning, and Pistis-Agentic, which additionally incorporates agentic trajectory data to support long-horizon planning, iterative reasoning, and tool use. Pistis-Agentic is particularly strong in multimodal search. Both scales outperform their corresponding base models. Beyond model-parameter optimization, we further introduce Pistis-Auto-Harnessing (PAH), a system-level method that automatically improves the agent's inference harness through iterative optimization. Experiments demonstrate that PAH enhances the model performance without updating the model parameters or increasing the interaction budget.

cs.AI↗
arXiv2026

Mind the Gap: Rethinking I/O Design for Contact-Rich Visuomotor Policy Learning

Contact-rich teleoperation logs expose a policy I/O design choice: demonstrations may contain the robot execution (E), leader command (C), or both. These signals are not interchangeable: E2E may discard contact-generating command offsets, whereas E2C preserves these offsets but omits the robot's execution response. We propose Dual-State Conditioning (EC2C), which conditions on both E and C while predicting future C, exposing command-execution mismatch as a cue for contact, latency, payload, and operator compensation; in quasi-static contact, this cue is often force-correlated. On a low-cost setup without force, tactile, or motor-current policy input, EC2C outperforms E2E and a strong E2C baseline across several real-world contact-rich, force-sensitive, and dynamic tasks. These results support EC2C as a practical default I/O setting for contact-rich imitation learning. We further formulate latency-adaptive inpainting as a temporal extension of this I/O choice for action-chunking policies, and discuss when long histories help dynamic inference or introduce causal confounding.

cs.RO↗
arXiv2026

LiDARDraft: Generating LiDAR Point Cloud from Versatile Inputs

Generating realistic and diverse LiDAR point clouds is crucial for autonomous driving simulation. Although previous methods achieve LiDAR point cloud generation from user inputs, they struggle to attain high-quality results while enabling versatile controllability, due to the imbalance between the complex distribution of LiDAR point clouds and the simple control signals. To address the limitation, we propose LiDARDraft, which utilizes the 3D layout to build a bridge between versatile conditional signals and LiDAR point clouds. The 3D layout can be trivially generated from various user inputs such as textual descriptions and images. Specifically, we represent text, images, and point clouds as unified 3D layouts, which are further transformed into semantic and depth control signals. Then, we employ a rangemap-based ControlNet to guide LiDAR point cloud generation. This pixel-level alignment approach demonstrates excellent performance in controllable LiDAR point clouds generation, enabling "simulation from scratch", allowing self-driving environments to be created from arbitrary textual descriptions, images and sketches.

cs.CV↗
arXiv2026

The Curse of Precision: A Data Scaling Law for High-Precision Robotic Manipulation

While scaling laws for imitation learning have primarily focused on generalization in open-world settings, the relationship between data and precision in closed-world tasks like robotic assembly remains largely unexplored. This paper systematically investigates this relationship and introduces a novel scaling law. We find that to achieve a fixed success rate, the required number of demonstrations $N$ grows super-exponentially as the target precision $P$ approaches a limit $c$. This relationship is accurately captured by the model $\log(N) \propto 1/(P-c)$. Crucially, we reveal that the limit precision $c$ is not a static physical constant of the task but an emergent property of the entire agent system, including its sensors and expert policy. Through experiments on canonical manipulation tasks, we validate this law and demonstrate that improving system components, such as adding a wrist camera or using a more effective expert, measurably lowers $c$, thus expanding the system's achievable precision. Our work provides a new theoretical framework for precision in robotics and a quantitative metric to evaluate system capabilities. Furthermore, these findings provide a practical methodology for guiding the development and debugging of high-precision manipulation systems.

cs.RO↗
arXiv2026

Dynamic Interaction-Aware and Causality-Disentangled Framework for Multimodal Sentiment Analysis

Although Multimodal Sentiment Analysis (MSA) effectively leverages rich information from language, visual, and acoustic modalities, existing methods still face two core challenges: 1) static conflict suppression mechanisms fail to adapt to dynamic variations across samples, and 2) the inherent sentimental bias within the language modality, which can misguide learning from other modalities, remains entangled. To this end, we propose a Dynamic Multimodal Causal Disentanglement and Adaptive Fusion Framework (MCAF). Its cornerstone is the Multi-Granularity Causal Dynamic Router and a Conditional Diffusion Denoising Module. First, we introduce a causal intervention module based on the information bottleneck principle, which builds a Structural Causal Model to disentangle sentimental bias from language features, yielding a "de-confounded" language representation as a pure guiding signal. Second, we devise a Dynamic Multimodal Router that evaluates the interaction states (complementary, conflicting, or redundant) among visual, acoustic, and de-confounded language signals in real-time across three levels: feature, temporal, and modality, then adaptively allocates weights and routes information flow for fine-grained regulation. Finally, a lightweight Conditional Diffusion Denoising Module performs iterative denoising on the fused joint representation to explicitly filter out residual irrelevant information, generating a robust hyper-modality representation. Extensive experiments on the CMU-MOSI and CMU-MOSEI benchmarks show that MCAF sets new state-of-the-art on key classification metrics, achieving an Acc-2/F1 of 86.52%/86.51% on MOSI and 86.72%/86.65% on MOSEI, while remaining highly competitive on others. Comprehensive analyses and visualizations further validate its efficacy in dynamically perceiving interactions, disentangling bias, and enhancing interpretability.

cs.MM↗
arXiv2026

Scalable unsupervised feature selection via weight stability

Unsupervised feature selection is critical for improving clustering performance in high-dimensional data, where irrelevant features can obscure meaningful structure. In this work, we propose the Minkowski weighted $k$-means++, a novel initialisation strategy for the Minkowski Weighted $k$-means. Our initialisation selects centroids probabilistically using feature relevance estimates derived from the data itself. Building on this, we propose two new feature selection algorithms, FS-MWK++, which aggregates feature weights across a range of Minkowski exponents identifying stable and informative features, and SFS-MWK++, a scalable variant based on subsampling. We support our approach with a theoretical analysis, demonstrating that, under explicit assumptions on noise features and cluster structure, relevant features are assigned consistently higher weights than noise features across a range of Minkowski exponents. Our software can be found at https://github.com/xzhang4-ops1/FSMWK.

cs.LG↗
arXiv2026

VLBM: Variational Latent Basis Modeling for OOD Robust Multivariate Time Series Forecasting

Out of distribution (OOD) events in multivariate time series forecasting are rare but often dominate real world risk, making average case forecasting insufficient for reliable deployment. Under standard average risk training on mixed ID/OOD distributions, optimization signals from rare OOD events can be overwhelmed by frequent in distribution (ID) patterns, so strong benchmark accuracy may not translate into reliability under high impact shifts. To address this issue, we propose VLBM (Variational Latent Basis Model), a theory guided latent forecasting framework that separates stable dynamics from OOD induced deviations. VLBM learns a shared latent basis that defines a low rank subspace for stable ID dynamics, explicitly decomposes inputs into basis subspace components and orthogonal residual components, and aligns a future aware posterior with a future blind prior so that test time latent inference depends only on historical input. Across 12 benchmark tasks spanning transportation, weather, power systems, and other real world domains, including newly constructed real world OOD traffic datasets, VLBM achieves state of the art OOD robustness and ID accuracy, with average MAE and MSE gains of 15.08\% and 7.74\% over the strongest baseline. On a synthetic simulation dataset, VLBM also consistently achieves the best performance and better tracks OOD pulse recovery. These results support latent structured forecasting as a principled route to robust prediction under mixed ID and OOD conditions. The code is available at https://github.com/leijieruilq/VLBM_OOD_forecast.

cs.LG↗
arXiv2026

The Sword, Shield, and Achilles' Heel: Characterizing the Linguistic Inductive Bias of Large Language Models for Spatial Reasoning in Navigation Planning

Large Language Model (LLM)-based navigation systems commonly construct explicit spatial representations (e.g., topological graphs, semantic raster maps) and translate them into textual descriptions as LLMs' inputs. However, the linguistic structures of such text-based spatial representations and the choices of contextual features (e.g., topology, geometry) they contain are often treated as neutral engineering decisions rather than key factors that shape LLMs' behavior. To fill the gap, we propose a dual-interventional framework that disentangles linguistic structures from different contextual cues to evaluate the linguistic inductive bias of LLMs for navigation planning. In the framework, representation intervention varies the linguistic format and the degree of linguistic compression, clarifying when linguistic representations support or inhibit navigation planning. Context intervention, combined with contextual feature combination and conflict probing, explicitly clarifies the preferences and weaknesses of LLMs when processing different contextual cues. Experiments across diverse spatial reasoning tasks and multiple model scales reveal a consistent pattern: topological information is a sturdy shield and the backbone of robust planning; linguistic format is a double-edged sword whose effect depends on model size, task demands, and the compression level; and semantic information is a fatal Achilles' heel -- incorrect semantic cues can systematically derail the planning process. Overall, our study shows that effective text-based spatial representations in LLM-based navigation should preserve topological integrity, calibrate representational compression to model capacity, and ensure semantic correctness, rather than simply adopting a single representation. Our code is publicly available at https://github.com/jonesdong150/LLM-Navigation-Inductive-Bias.

cs.CL↗
arXiv2026

Global Policy-Space Response Oracles for Two-Player Zero-Sum Games

The Policy-Space Response Oracles (PSRO) framework scales equilibrium computation to large zero-sum games by iteratively expanding a restricted strategy set using deep reinforcement learning (DRL). A central challenge is to construct, under limited computational budgets, a small strategy population whose induced game well approximates the full game. Existing PSRO variants typically expand the population using best responses to meta-strategies computed from restricted-game payoffs, which can lead to inefficient expansions that provide limited global improvement. We propose to guide population expansion by directly evaluating the post-expansion population quality. Specifically, we adopt Population Exploitability (PE) to measure how well a restricted strategy set represents the full game, and introduce a two-phase exploration--selection framework that explicitly minimizes PE during expansion. We instantiate this framework as Global PSRO, a practical DRL-based algorithm that efficiently generates candidate responses and estimates PE via parameter-sharing conditional neural networks. Experiments across multiple two-player zero-sum games show that Global PSRO achieves lower exploitability and approximates Nash equilibria with significantly fewer policy iterations than prior PSRO methods.

cs.AI↗
arXiv2026

Accelerating Black Hole Image Generation via Latent Space Diffusion Models

Interpreting horizon-scale black hole images currently relies on computationally intensive General Relativistic Ray Tracing (GRRT) simulations, which pose a significant bottleneck for rapid parameter exploration and high-precision tests of strong-field gravity. We demonstrate that physically accurate black hole images, synthesized from magnetized accretion flows, inherently reside on a low-dimensional manifold-encoding the essential features of spacetime geometry, plasma distribution, and relativistic emission. Leveraging this structure, we introduce a physics-conditioned diffusion model that operates in a compact latent space to generate high-fidelity black hole imagery directly from physical parameters. The model accurately reproduces critical observational signatures from full GRRT simulations-such as shadow diameter, photon-ring structure, and relativistic brightness asymmetry-while achieving over a fourfold reduction in computational expense. Compared with the previous generation of denoising diffusion models, the proposed approach achieves significant improvements in image quality, reconstruction fidelity, and parameter estimation accuracy, while reducing the average inference time per black hole image from 5.25 seconds to 1.15 seconds. Our work establishes diffusion-based latent models as efficient and scalable substitutes for traditional radiative transfer solvers, offering a practical framework toward real-time modeling and inference for next-generation black hole imaging.

gr-qc↗
arXiv2025

Unified Bulk-Entanglement Correspondence in Non-Hermitian Systems

The non-Hermitian skin effect (NHSE) fundamentally invalidates the conventional bulk-boundary correspondence (BBC), leading topological diagnostics into a crisis. While the non-Bloch polarization $P_β$ defined on the generalized Brillouin zone restores momentum-space topology, a direct, robust real-space bulk probe has remained elusive. We resolve this by establishing a universal correspondence between $P_β$ and the entanglement polarization $χ$ of the biorthogonal ground state. Introducing a quasi-reciprocal Hamiltonian $\tilde{H}$ that removes the NHSE while preserving bulk topology, we rigorously prove the fundamental identity $P_β \equiv χ(\tilde{H})\pmod 1$ in the thermodynamic limit under the quasi-locality assumption. Crucially, we demonstrate that this equivalence transcends the locality constraints that limit traditional topological invariants. While the conventional Resta polarization fails when $\tilde{H}$ becomes non-local due to the divergence of position variance, we reveal that $χ(\tilde{H})$ remains robustly quantized, protected by the Fredholm index of Toeplitz operators. Our work thus identifies entanglement as the unique real-space diagnostic capable of capturing non-Bloch topology beyond the breakdown of locality, successfully restoring the BBC across diverse non-Hermitian systems such as line-gap, point-gap, and gapless phases, thereby unifying the geometric and entanglement paradigms in non-Hermitian physics.

quant-ph↗