arXiv subjects

James Wang

Publications and source records attributed to James Wang.

At least 19 recordsLinked to original sources

arXiv2026

Scalable Online Flight Trajectory Optimization via Sequential Quadratic Programming for Urban Air Mobility in Ultra Low-Altitude Airspace

As Urban Air Mobility (UAM) scales toward high-density operations, generating collision-free trajectories within complex 3D cityscapes is a critical safety requirement. This paper proposes a scalable Sequential Quadratic Programming (SQP) framework that integrates geometric environmental constraints, operational limits, and vehicle dynamics within a single online trajectory optimization process. Rather than precomputing obstacle-free corridors ahead of time, our method encodes obstacle avoidance as live separating-hyperplane constraints regenerated at every solver iteration, so that dense urban geometry and full-DOF vehicle dynamics are resolved jointly and online as the reference and environment evolve. A variable-scale quadtree decomposition keeps computation bounded, enabling the framework to scale to city-wide environments while preserving real-time performance for high-speed flight. We validate the framework against conventional SQP, Iterative Linear Quadratic Regulator, and Differential Dynamic Programming across flights in five real-world urban centers, attaining 100% success and clearance rates on CPU-only hardware.

eess.SY↗

arXiv2026

Suppressing Intrinsic Spin-Phonon Errors in Trapped-Ion Quantum Simulation

Trapped-ion quantum simulators realize programmable spin models through phonon-mediated interactions. For Hamiltonians with noncommuting terms, however, the same phonon bus generates intrinsic spin-phonon errors that strongly distort the target dynamics. Because these errors are governed by the full time history of the spin-dependent phonon motion, they survive standard loop-closing control and limit simulation accuracy. Using a sequence of frame transformations, we isolate the residual error dynamics and show that this intrinsic error can be strongly suppressed while preserving programmable Ising couplings. Full spin-boson simulations of multi-ion chains demonstrate orders-of-magnitude lower error than both constant-drive and conventional loop-closing protocols. These results remove a central precision barrier in trapped-ion analog quantum simulation and enable accurate programmable simulation of noncommuting many-body Hamiltonians and dynamical protocols.

quant-ph↗

arXiv2026

WALL-WM: Carving World Action Modeling at the Event Joints

WALL-WM is a World Action Model that shifts video-action learning from chunk-centric optimization to event-grounded Vision-Language-Action pretraining, using semantically coherent action events as the atomic unit of learning. Existing WAMs commonly initialize from multimodal or video foundation models and then optimize fixed-length action chunks conditioned directly on the current observation and instruction. Although convenient, this chunk-centric formulation creates a fundamental granularity mismatch. Language describes semantic goals and events, vision evolves through continuous scene dynamics, and actions operate at control-level timescales; forcing all three into the same fixed-length prediction window turns VLA training into short-horizon correlation fitting. WALL-WM addresses this mismatch by organizing both supervision and data around semantic events. Specifically, it pairs event-grounded VLA pretraining with a data ecosystem built from event-level captions and cluster-balanced sampling, enabling scalable learning over diverse behaviors, scenes, and task structures. From the same event-pretrained backbone, WALL-WM supports two complementary inference modes. The event mode consumes next-event descriptions and enables variable-length execution chunks, while the unified mode uses a VLM with Staircase Decoding to condition conventional fixed-length chunk inference while preserving a gradient-continuous VLA path. Together with Muon-optimizer-based large-scale pretraining infrastructure, WALL-WM provides a practical scale-up recipe for general-purpose WAMs. Experiments show that WALL-WM generalizes broadly across language, scenes, and tasks, achieving state-of-the-art performance in large-scale real-world generalization evaluation.

cs.RO↗

arXiv2026

Wall-OSS-0.5 Technical Report

Large-scale Vision-Language-Action (VLA) pretraining is increasingly adopted as the foundation for robot policies, yet the evidence for pretrained VLAs is almost invariably reported after task-specific fine-tuning. This leaves a foundational question unanswered: does VLA pretraining itself yield executable robot behavior, or does it merely furnish a better initialization for downstream policy learning? We present Wall-OSS-0.5, an open-source 4B VLA built upon a 3B VLM backbone augmented with action-generation components, designed so that pretrained robotic capability is directly measurable on physical hardware. The model is pretrained across more than 20 embodiments, processing over one million robot trajectories per epoch alongside a grounded multimodal corpus. We adopt a gradient-bridged co-training recipe in which three objectives play distinct and complementary roles: discrete action prediction routes strong VLM-native gradients into the backbone, multimodal prediction preserves grounded vision-language understanding, and continuous flow matching serves as the deployment-time action interface. Before task-specific fine-tuning, the pretrained checkpoint achieves non-trivial zero-shot real-robot behavior, completing several tasks, including a held-out deformable manipulation task, at high task progress on a 17-task suite. After fine-tuning, the same checkpoint serves as a stronger adaptation prior, reaching 60.5% average task progress on 15 real-robot tasks and outperforming \pi_0.5 by 17.5%. Multimodal evaluations further confirm that action training does not erode grounded vision-language competence: the model preserves broad vision-language ability while strengthening embodied grounding. Together, these results reposition VLA pretraining from an initialization strategy to a directly testable, already useful source of robot capability.

cs.RO↗

arXiv2026

18-dB on-chip vacuum squeezing in an adaptively poled lithium niobate waveguide

Quantum squeezed states of light can enhance measurement sensitivity beyond classical limits and enable quantum information processing, but scalable low-loss sources remain challenging. We demonstrate continuous-wave quantum squeezing on a chip, achieving 18 dB of squeezing and 20 dB of anti-squeezing at 1570 nm in a 1.6-cm traveling-wave adaptively poled thin-film lithium niobate waveguide. A distributed model independently determines facet losses, phase noise, and nonlinear interaction strength without prior assumptions, enabling rigorous inference of on-chip performance. We estimate a 95% confidence interval of [-18.96, -17.25] dB squeezing and [19.96, 21.35] dB anti-squeezing. These values represent the highest squeezing reported for any integrated photonic platform and the first assumption-free statistical validation of integrated squeezing performance. Our results establish thin-film lithium niobate as a high-performance, scalable platform for continuous-variable quantum sensing, communications, and photonic computing.

physics.optics↗

arXiv2026

Learning When to Stop: Selective Imitation Learning Under Arbitrary Dynamics Shift

Behavior cloning provides strong imitation learning guarantees when training and test environments share the same dynamics. However, in many deployment settings the test environment's transitions differ from training, and classical offline IL offers no recourse: the learner must commit to an action at every state, even when its demonstrations are uninformative and could lead to arbitrary degradation of performance. This motivates the study of selective imitation, where the learner may choose to stop when it cannot act reliably. We introduce a model for selective imitation under arbitrary dynamics shift: given labeled expert demonstrations from a training environment and unlabeled state trajectories from the same expert in a test environment, the learner outputs a selective policy that is complete (rarely stops in training) and sound (incurs low regret before stopping in test). Our algorithm, SeqRejectron, constructs a stopping rule using a small set of validator policies whose size is independent of the horizon or policy class. For deterministic policies, this yields horizon-free $\tilde{O}(\log|\Pi|/\epsilon^2)$ sample complexity, assuming sparse costs. For stochastic policies, we obtain analogous horizon-free guarantees using a cumulative Hellinger stopping time. We extend the framework to misspecified experts and different expert policies across train and test and obtain results that gracefully degrade with the amount of misspecification.

cs.LG↗

arXiv2026

Weight Clipping for Robust Conformal Inference under Unbounded Covariate Shifts

Conformal prediction (CP) provides powerful, distribution-free prediction sets, but its guarantees rely on the exchangeability of training and test data, which is often violated in practice due to covariate shifts. While weighted conformal prediction (WCP) is designed to handle such shifts, it can suffer from significant undercoverage when the density ratio between the distributions is unbounded and/or must be learned. This is because of both overfitting in learning the density ratio, and high variance in estimating the nonconformity score threshold. To address this, we introduce clipped least-squares importance fitting (CLISF) as a reduced-variance method for density ratio estimation. Specifically, we show that density ratios learned using CLISF, when plugged into WCP, have bounded expected undercoverage. Furthermore, we show that the undercoverage can be corrected by running WCP with a slightly inflated coverage target; crucially, we are able to estimate the required level of inflation from the data. We provide the first theoretical guarantees for weight clipping in conformal inference, achieving dataset-conditional coverage with a sample complexity that does not blow up with the higher moments of the true density ratio -- a key limitation of prior work. We verify our results on real-world benchmarks and synthetic data.

cs.LG↗

arXiv2026

XRZero-G0: Pushing the Frontier of Dexterous Robotic Manipulation with Interfaces, Quality and Ratios

The acquisition of high-quality, action-aligned demonstration data remains a fundamental bottleneck in scaling foundation models for dexterous robot manipulation. Although robot-free human demonstrations (e.g., the UMI paradigm) offer a scalable alternative to traditional teleoperation, current systems are constrained by sub-optimal hardware ergonomics, open-loop workflows, and a lack of systematic data-mixing strategies. To address these limitations, we present XRZero-G0, a hardware-software co-designed system for embodied data collection and policy learning. The system features an ergonomic, virtual reality interface equipped with a top-view camera and dual specialized grippers to directly improve collection efficiency. To ensure dataset reliability, we propose a closed-loop collection, inspection, training, and evaluation pipeline for non-proprioceptive data. This workflow achieves an 85% data validity rate and establishes a transparent mechanism for quality control. Furthermore, we investigate the empirical scaling behaviors and optimal mixing ratios of robot-free data. Extensive experiments indicate that combining a minimal volume of real-robot data with large-scale robot-free data (e.g., a 10:1 ratio) achieves performance comparable to exclusively real-robot datasets, while reducing acquisition costs by a factor of twenty. Utilizing XRZero-G0, we construct a 2,000-hour robot-free dataset that enables zero-shot cross-embodiment transfer to a target physical robot, demonstrating a highly scalable methodology for generalized real-world manipulation.Our project repository: https://github.com/X-Square-Robot/XRZero-G0

cs.RO↗

arXiv2026

Moonshine v2: Ergodic Streaming Encoder ASR for Latency-Critical Speech Applications

Latency-critical speech applications (e.g., live transcription, voice commands, and real-time translation) demand low time-to-first-token (TTFT) and high transcription accuracy, particularly on resource-constrained edge devices. Full-attention Transformer encoders remain a strong accuracy baseline for automatic speech recognition (ASR) because every frame can directly attend to every other frame, which resolves otherwise locally ambiguous acoustics using distant lexical context. However, this global dependency incurs quadratic complexity in sequence length, inducing an inherent "encode-the-whole-utterance" latency profile. For streaming use cases, this causes TTFT to grow linearly with utterance length as the encoder must process the entire prefix before any decoder token can be emitted. To better meet the needs of on-device, streaming ASR use cases we introduce Moonshine v2, an ergodic streaming-encoder ASR model that employs sliding-window self-attention to achieve bounded, low-latency inference while preserving strong local context. Our models achieve state of the art word error rates across standard benchmarks, attaining accuracy on-par with models 6x their size while running significantly faster. These results demonstrate that carefully designed local attention is competitive with the accuracy of full attention at a fraction of the size and latency cost, opening new possibilities for interactive speech interfaces on edge devices.

cs.CL↗

arXiv2026

A Migration-Assisted Deep Learning Scheme for Imaging Defects Inside Cylindrical Structures via GPR: A Case Study for Tree Trunks

Ground-penetrating radar (GPR) has emerged as a prominent tool for imaging internal defects in cylindrical structures, such as columns, utility poles, and tree trunks. However, accurately reconstructing both the shape and permittivity of the defects inside cylindrical structures remains challenging due to complex wave scattering phenomena and the limited accuracy of the existing signal processing and deep learning techniques. To address these issues, this study proposes a migration-assisted deep learning scheme for reconstructing the shape and permittivity of defects within cylindrical structures. The proposed scheme involves three stages of GPR data processing. First, a dual-permittivity estimation network extracts the permittivity values of the defect and the cylindrical structure, the latter of which is estimated with the help of a novel structural similarity index measure-based autofocusing technique. Second, a modified Kirchhoff migration incorporating the extracted permittivity of the cylindrical structure maps the signals reflected from the defect to the imaging domain. Third, a shape reconstruction network processes the migrated image to recover the precise shape of the defect. The image of the interior defect is finally obtained by combining the reconstructed shape and extracted permittivity of the defect. The proposed scheme is validated using both synthetic and experimental data from a laboratory trunk model and real tree trunk samples. Comparative results show superior performance over existing deep learning methods, while generalization tests on live trees confirm its feasibility for in-field deployment. The underlying principle can further be applied to other circumferential GPR imaging scenarios. The code and database are available at: https://github.com/jwqian54/Migration-Assisted-DL.

eess.SP↗

arXiv2025

MiroThinker: Pushing the Performance Boundaries of Open-Source Research Agents via Model, Context, and Interactive Scaling

We present MiroThinker v1.0, an open-source research agent designed to advance tool-augmented reasoning and information-seeking capabilities. Unlike previous agents that only scale up model size or context length, MiroThinker explores interaction scaling at the model level, systematically training the model to handle deeper and more frequent agent-environment interactions as a third dimension of performance improvement. Unlike LLM test-time scaling, which operates in isolation and risks degradation with longer reasoning chains, interactive scaling leverages environment feedback and external information acquisition to correct errors and refine trajectories. Through reinforcement learning, the model achieves efficient interaction scaling: with a 256K context window, it can perform up to 600 tool calls per task, enabling sustained multi-turn reasoning and complex real-world research workflows. Across four representative benchmarks-GAIA, HLE, BrowseComp, and BrowseComp-ZH-the 72B variant achieves up to 81.9%, 37.7%, 47.1%, and 55.6% accuracy respectively, surpassing previous open-source agents and approaching commercial counterparts such as GPT-5-high. Our analysis reveals that MiroThinker benefits from interactive scaling consistently: research performance improves predictably as the model engages in deeper and more frequent agent-environment interactions, demonstrating that interaction depth exhibits scaling behaviors analogous to model size and context length. These findings establish interaction scaling as a third critical dimension for building next-generation open research agents, complementing model capacity and context windows.

cs.CL↗

arXiv2025

High Clockrate Free-space Optical In-Memory Computing

The ability to process and act on data in real time is increasingly critical for applications ranging from autonomous vehicles, three-dimensional environmental sensing and remote robotics. However, the deployment of deep neural networks (DNNs) in edge devices is hindered by the lack of energy-efficient scalable computing hardware. Here, we introduce a fanout spatial time-of-flight optical neural network (FAST-ONN) that calculates billions of convolutions per second with ultralow latency and power consumption. This is enabled by the combination of high-speed dense arrays of vertical-cavity surface-emitting lasers (VCSELs) for input modulation with spatial light modulators of high pixel counts for in-memory weighting. In a three-dimensional optical system, parallel differential readout allows signed weight values accurate inference in a single shot. The performance is benchmarked with feature extraction in You-Only-Look-Once (YOLO) for convolution at 100 million frames per second (MFPS), and in-system backward propagation training with photonic reprogrammability. The VCSEL transmitters are implementable in any free-space optical computing systems to improve the clockrate to over gigahertz. The high scalability in device counts and channel parallelism enables a new avenue to scale up free space computing hardware.

cs.ET↗

arXiv2025

Igniting VLMs toward the Embodied Space

While foundation models show remarkable progress in language and vision, existing vision-language models (VLMs) still have limited spatial and embodiment understanding. Transferring VLMs to embodied domains reveals fundamental mismatches between modalities, pretraining distributions, and training objectives, leaving action comprehension and generation as a central bottleneck on the path to AGI. We introduce WALL-OSS, an end-to-end embodied foundation model that leverages large-scale multimodal pretraining to achieve (1) embodiment-aware vision-language understanding, (2) strong language-action association, and (3) robust manipulation capability. Our approach employs a tightly coupled architecture and multi-strategies training curriculum that enables Unified Cross-Level CoT-seamlessly unifying instruction reasoning, subgoal decomposition, and fine-grained action synthesis within a single differentiable framework. Our results show that WALL-OSS attains high success on complex long-horizon manipulations, demonstrates strong instruction-following capabilities, complex understanding and reasoning, and outperforms strong baselines, thereby providing a reliable and scalable path from VLMs to embodied foundation models.

cs.RO↗

arXiv2025

Flavors of Moonshine: Tiny Specialized ASR Models for Edge Devices

We present the Flavors of Moonshine, a suite of tiny automatic speech recognition (ASR) models specialized for a range of underrepresented languages. Prevailing wisdom suggests that multilingual ASR models outperform monolingual counterparts by exploiting cross-lingual phonetic similarities. We challenge this assumption, showing that for sufficiently small models (27M parameters), training monolingual systems on a carefully balanced mix of high-quality human-labeled, pseudo-labeled, and synthetic data yields substantially superior performance. On average, our models achieve error rates 48% lower than the comparably sized Whisper Tiny model, outperform the 9x larger Whisper Small model, and in most cases match or outperform the 28x larger Whisper Medium model. These results advance the state of the art for models of this size, enabling accurate on-device ASR for languages that previously had limited support. We release Arabic, Chinese, Japanese, Korean, Ukrainian, and Vietnamese Moonshine models under a permissive open-source license.

cs.CL↗

arXiv2025

Exploring Partial Multi-Label Learning via Integrating Semantic Co-occurrence Knowledge

Partial multi-label learning aims to extract knowledge from incompletely annotated data, which includes known correct labels, known incorrect labels, and unknown labels. The core challenge lies in accurately identifying the ambiguous relationships between labels and instances. In this paper, we emphasize that matching co-occurrence patterns between labels and instances is key to addressing this challenge. To this end, we propose Semantic Co-occurrence Insight Network (SCINet), a novel and effective framework for partial multi-label learning. Specifically, SCINet introduces a bi-dominant prompter module, which leverages an off-the-shelf multimodal model to capture text-image correlations and enhance semantic alignment. To reinforce instance-label interdependencies, we develop a cross-modality fusion module that jointly models inter-label correlations, inter-instance relationships, and co-occurrence patterns across instance-label assignments. Moreover, we propose an intrinsic semantic augmentation strategy that enhances the model's understanding of intrinsic data semantics by applying diverse image transformations, thereby fostering a synergistic relationship between label confidence and sample difficulty. Extensive experiments on four widely-used benchmark datasets demonstrate that SCINet surpasses state-of-the-art methods.

cs.CV↗

arXiv2025

Synthetic CT Generation from Time-of-Flight Non-Attenutaion-Corrected PET for Whole-Body PET Attenuation Correction

Positron Emission Tomography (PET) imaging requires accurate attenuation correction (AC) to account for photon loss due to tissue density variations. In PET/MR systems, computed tomography (CT), which offers a straightforward estimation of AC is not available. This study presents a deep learning approach to generate synthetic CT (sCT) images directly from Time-of-Flight (TOF) non-attenuation corrected (NAC) PET images, enhancing AC for PET/MR. We first evaluated models pre-trained on large-scale natural image datasets for a CT-to-CT reconstruction task, finding that the pre-trained model outperformed those trained solely on medical datasets. The pre-trained model was then fine-tuned using an institutional dataset of 35 TOF NAC PET and CT volume pairs, achieving the lowest mean absolute error (MAE) of 74.49 HU and highest peak signal-to-noise ratio (PSNR) of 28.66 dB within the body contour region. Visual assessments demonstrated improved reconstruction of both bone and soft tissue structures from TOF NAC PET images. This work highlights the effectiveness of using pre-trained deep learning models for medical image translation tasks. Future work will assess the impact of sCT on PET attenuation correction and explore additional neural network architectures and datasets to further enhance performance and practical applications in PET imaging.

eess.IV↗

arXiv2025

Logarithmic Regret for Nonlinear Control

We address the problem of learning to control an unknown nonlinear dynamical system through sequential interactions. Motivated by high-stakes applications in which mistakes can be catastrophic, such as robotics and healthcare, we study situations where it is possible for fast sequential learning to occur. Fast sequential learning is characterized by the ability of the learning agent to incur logarithmic regret relative to a fully-informed baseline. We demonstrate that fast sequential learning is achievable in a diverse class of continuous control problems where the system dynamics depend smoothly on unknown parameters, provided the optimal control policy is persistently exciting. Additionally, we derive a regret bound which grows with the square root of the number of interactions for cases where the optimal policy is not persistently exciting. Our results provide the first regret bounds for controlling nonlinear dynamical systems depending nonlinearly on unknown parameters. We validate the trends our theory predicts in simulation on a simple dynamical system.

cs.LG↗

arXiv2024

Moonshine: Speech Recognition for Live Transcription and Voice Commands

This paper introduces Moonshine, a family of speech recognition models optimized for live transcription and voice command processing. Moonshine is based on an encoder-decoder transformer architecture and employs Rotary Position Embedding (RoPE) instead of traditional absolute position embeddings. The model is trained on speech segments of various lengths, but without using zero-padding, leading to greater efficiency for the encoder during inference time. When benchmarked against OpenAI's Whisper tiny-en, Moonshine Tiny demonstrates a 5x reduction in compute requirements for transcribing a 10-second speech segment while incurring no increase in word error rates across standard evaluation datasets. These results highlight Moonshine's potential for real-time and resource-constrained applications.

cs.SD↗