Searcharxiv⌕ Search

arXiv subjects

Long Qian

Publications and source records attributed to Long Qian.

At least 19 recordsLinked to original sources
arXiv2026

Messages, Not Tokens: Grounded Coresets for Faithful VLM Compression

Modern vision language models (VLMs) turn high-resolution images into long sequences of visual tokens. Every token traverses the language decoder and persists in its prompt KV cache, inflating inference cost and motivating aggressive visual compression. Existing score-based methods assign each token an independent importance score and retain the Top-K. However, text queries consume collective, signed attention messages from the visual population, not isolated patches. Consequently, equally sized Top-K sets can repeatedly cover one salient region, omit sparse but complementary evidence and discard information carried by the removed population. We therefore formulate faithful visual compression as constructing a compact coreset for decoder messages, and introduce our training-free Grounded Message Coreset Pruning (GMC) which jointly allocates support across query-grounded, appearance, and coordinate-aware evidence, then transports discarded states into selected representatives at their original multimodal positions before physical compaction and native attention resume. This decomposes faithful compression into two coupled components, including selecting carriers that cover the required message modes and realizing the signed population message on those carriers. We further derive bounds connecting their errors to signed-message distortion, visual innovation, and candidate-margin stability. Experiments across multiple VLM families and diverse benchmarks demonstrate strong performance, with GMC-H2 retaining 97.78% Full-relative mean capability on Qwen2.5-VL-7B using 80.2% fewer visual tokens, while GMC-L16 reaches 100.36%. Controlled interventions verify that collective support and population realization jointly drive these gains.

cs.CV↗
arXiv2026

Xiaomi-Robotics-U0: Unified Embodied Synthesis with World Foundation Model

Recent foundation image and video generation models offer strong generalization and controllability, but their direct application to embodied scenarios is limited by requirements for multi-view consistency, geometric coherence, and robot embodiment constraints. Existing methods typically adapt foundation models with limited robot data, often sacrificing visual knowledge acquired during large-scale pre-training. We present Xiaomi-Robotics-U0, a 38-billion-parameter multimodal autoregressive model for unified embodied synthesis. It treats embodied generation as an extension of foundation image and video generation and jointly optimizes text-to-image generation, image editing, embodied scene generation, embodied transfer, and embodied video generation. This unified framework preserves the generalization of the pre-trained world foundation model while adapting it to embodied settings. Xiaomi-Robotics-U0 is the first model to support high-quality multi-view scene generation across multiple robot embodiments and to introduce structured, controllable embodied transfer for fine-grained editing while preserving multi-view consistency and interaction dynamics. It achieves state-of-the-art results on single-step and sequential generation tasks, outperforming GPT-Image-2.0 in human evaluations of embodied scene generation and transfer, ranking first on World Arena for embodied video generation, and improving the out-of-distribution success rate of pi_0.5 from 36.9% to 63.2% on challenging real-world manipulation tasks. These results show that foundation world models can serve both as embodied world models and scalable data engines for embodied intelligence. Code and checkpoints are available at https://robotics.xiaomi.com/xiaomi-robotics-u0.html.

cs.RO↗
arXiv2026

Differential Equation Inductive Robustness Axiomatization

This article establishes the completeness of an axiomatization for the robust safety of dynamical systems with polynomial differential equations on bounded time horizons. Safety properties of robust systems are uniformly reduced to a sound axiomatization of polynomial invariants, resulting in reliable logical proofs of correctness. Approximate decidability results are also established: there is a computable algorithm such that, given any perturbation parameter $\delta$, it either produces a symbolic proof of robust safety (hence correctly decides the dynamical system to be robustly safe), or correctly decides that the system is not robustly safe under a perturbation of level $\delta$. In contrast to earlier works, this article crucially leverages results from subanalytic geometry to retain a level of exactness, thereby establishing positive results of provability/decidability allowing for arbitrary bounded (semialgebraic) initial/post conditions even without positive separation at their (topological) boundaries. This enables the generation of proofs of inductive safety beyond finite time horizons for general hybrid dynamical systems.

cs.LO↗
arXiv2026

A Deductive Refinement Calculus for Differential-Algebraic Programs

This paper presents differential-algebraic refinement logic (dARL) with which one can deductively verify both properties and relations of differential-algebraic programs (DAPs) that extend hybrid dynamical systems with differential-algebraic equations (DAEs). A refinement calculus is introduced that enables the sound comparison of trajectories of differential-algebraic equations, crucially utilizing a novel trace-based semantics. This enables the incremental verification/simplification of complicated DAEs, while ensuring correctness at each step by the soundness of the calculus. The calculus is shown to be complete for certifying index reductions of DAEs, providing trustworthy syntactic proofs of correctness at each step of the reduction.

cs.LO↗
arXiv2026

SpatialVAM:Spatial-Aware Multi-View Video Diffusion as a Data-Efficient Robot Policy

Robotic manipulation requires understanding both the 3D spatial structure of the environment and its temporal evolution, yet most existing policies neglect one or both aspects. They often rely on 2D visual observations or backbones pretrained on static image--text pairs, which leads to high data requirements and limited comprehension of environment dynamics. To address this, we introduce SpatialVAM, the first 3D Video Action Model that simultaneously predict spatial-aware multi-view heatmap videos and RGB videos. Our key insight is that this design naturally injects 3D information into video foundation models while aligning the representation format between video pretraining and action finetuning. Extensive experiments demonstrate that SpatialVAM enables data-efficient, robust, generalizable, and interpretable manipulation. With only ten demonstration trajectories and no additional pretraining, SpatialVAM handles challenging long-horizon and contact-rich tasks, generalizes to out-of-distribution settings, and predicts realistic future videos. Evaluations on Meta-World (22\%$\uparrow$), RoboCasa (15\%$\uparrow$) and real-world robotic platforms (16\%$\uparrow$) show that SpatialVAM consistently outperforms other video action models, vision language action models and 3D-based policies, establishing a new state-of-the-art in data-efficient multi-task manipulation.

cs.RO↗
arXiv2026

Trident: Adaptive Scheduling for Heterogeneous Multimodal Data Pipelines

The rapid adoption of large language models and multimodal foundation models has made multimodal data preparation pipelines critical AI infrastructure. These pipelines interleave CPU-heavy preprocessing with accelerator-backed (GPU/NPU/TPU) inference and produce massive intermediate artifacts. Achieving high throughput is difficult because workloads are highly non-stationary: regime shifts, input-dependent inference, and transient memory spikes cause rapid performance fluctuations and out-of-memory (OOM) failures. Existing schedulers typically rely on threshold-based autoscaling or assume synchronous, homogeneous operators, leading to poor efficiency. We present Trident, an adaptive scheduling framework for heterogeneous multimodal pipelines on fixed-resource clusters. Trident closes the loop across three coupled layers: (i) an observation layer that estimates per-operator sustainable throughput for asynchronous operators via Gaussian Process regression with anomaly filtering; (ii) an adaptation layer that detects workload shifts online and performs memory-constrained Bayesian optimization to recommend OOM-safe configurations; and (iii) a scheduling layer that solves a mixed-integer linear program to jointly optimize operator parallelism, placement, and configuration transitions under heterogeneous compute and bandwidth constraints, accounting for cold-start overhead via rolling updates. Decisions trigger sample invalidation and model refresh to keep estimates consistent with the active configuration. Implemented on Ray Data, Trident improves end-to-end throughput by up to 2.01x on a document curation (PDF) pipeline and 1.88x on a video curation pipeline over a static baseline, with low overhead suitable for online re-optimization.

cs.DC↗
arXiv2026

Microwave vortex beam lasing via photonic time crystals

Microwave lasing carrying orbital angular momentum (OAM) holds significant potential for advanced applications in fields such as high-capacity communications, precision sensing, and radar imaging. However, conventional approaches to masers fail to produce emission with embedded OAM. The recent emergence of photonic time crystals (PTCs)-artificially structured media with periodically varying electromagnetic properties in time-offers a paradigm shift toward resonance-free lasing without the need for gain media. Yet, pioneering PTC designs have been based on three-dimensional bulk structures, which lack a surface-emitting configuration, and do not possess the capability to modulate OAM, thus hindering the realization of surface-emitted PTC masing that carries OAM. Here, we report the first experimental demonstration of non-resonant, gain medium-free, and surface-emitted microwave vortex beam lasing OAM using ring-shaped PTCs. By developing a multiplier-driven time-varying metamaterial that achieves over 100% equivalent permittivity modulation depth, we establish momentum bandgaps (k gaps) with sufficient bandwidth to overcome intrinsic losses and enable self-sustained coherent microwave amplification. Furthermore, space-time modulation induces non-reciprocity between clockwise and counterclockwise k gap modes within the circularly symmetric PTC structure, facilitating the selective generation of microwave lasing carrying OAM-a capability beyond the reach of conventional maser technologies. Our work bridges PTC physics with coherent OAM-carrying microwave emission, establishing a transformative platform for next-generation wireless communications, advanced sensing systems, and OAM-based technologies.

physics.optics↗
arXiv2025

SmoothGuard: Defending Multimodal Large Language Models with Noise Perturbation and Clustering Aggregation

Multimodal large language models (MLLMs) have achieved impressive performance across diverse tasks by jointly reasoning over textual and visual inputs. Despite their success, these models remain highly vulnerable to adversarial manipulations, raising concerns about their safety and reliability in deployment. In this work, we first generalize an approach for generating adversarial images within the HuggingFace ecosystem and then introduce SmoothGuard, a lightweight and model-agnostic defense framework that enhances the robustness of MLLMs through randomized noise injection and clustering-based prediction aggregation. Our method perturbs continuous modalities (e.g., images and audio) with Gaussian noise, generates multiple candidate outputs, and applies embedding-based clustering to filter out adversarially influenced predictions. The final answer is selected from the majority cluster, ensuring stable responses even under malicious perturbations. Extensive experiments on POPE, LLaVA-Bench (In-the-Wild), and MM-SafetyBench demonstrate that SmoothGuard improves resilience to adversarial attacks while maintaining competitive utility. Ablation studies further identify an optimal noise range (0.1-0.2) that balances robustness and utility.

cs.LG↗
arXiv2025

From Zonotopes to Proof Certificates: A Formal Pipeline for Safe Control Envelopes

Synthesizing controllers that enforce both safety and actuator constraints is a central challenge in the design of cyber-physical systems. State-of-the-art reachability methods based on zonotopes deliver impressive scalability, yet no zonotope reachability tool has been formally verified and the lack of end-to-end correctness undermines the confidence in their use for safety-critical systems. Although deductive verification with the hybrid system prover KeYmaera X could, in principle, resolve this assurance gap, the high-dimensional set representations required for realistic control envelopes overwhelm its reasoning based on quantifier elimination. To address this gap, we formalize how control-invariant sets serve as sound safety certificates. Building on that foundation, we develop a verification pipeline for control envelopes that unites scalability and formal rigor. First, we compute control envelopes with high-performance reachability algorithms. Second, we certify every intermediate result using provably correct logical principles. To accelerate this certification, we offload computationally intensive zonotope containment tasks to efficient numerical backends, which return compact witnesses that KeYmaera X validates rapidly. We show the practical utility of our approach through representative case studies.

cs.LO↗
arXiv2025

Quality-Aware Language-Conditioned Local Auto-Regressive Anomaly Synthesis and Detection

Despite substantial progress in anomaly synthesis methods, existing diffusion-based and coarse inpainting pipelines commonly suffer from structural deficiencies such as micro-structural discontinuities, limited semantic controllability, and inefficient generation. To overcome these limitations, we introduce ARAS, a language-conditioned, auto-regressive anomaly synthesis approach that precisely injects local, text-specified defects into normal images via token-anchored latent editing. Leveraging a hard-gated auto-regressive operator and a training-free, context-preserving masked sampling kernel, ARAS significantly enhances defect realism, preserves fine-grained material textures, and provides continuous semantic control over synthesized anomalies. Integrated within our Quality-Aware Re-weighted Anomaly Detection (QARAD) framework, we further propose a dynamic weighting strategy that emphasizes high-quality synthetic samples by computing an image-text similarity score with a dual-encoder model. Extensive experiments across three benchmark datasets-MVTec AD, VisA, and BTAD, demonstrate that our QARAD outperforms SOTA methods in both image- and pixel-level anomaly detection tasks, achieving improved accuracy, robustness, and a 5 times synthesis speedup compared to diffusion-based alternatives. Our complete code and synthesized dataset will be publicly available.

cs.CV↗
arXiv2025

Approximate Axiomatization for Differentially-Defined Functions

This article establishes a complete approximate axiomatization for the real-closed field $\mathbb{R}$ expanded with all differentially-defined functions, including special functions such as $\sin(x), \cos(x), e^x, \dots$. Every true sentence is provable up to some numerical approximation, and the truth of such approximations converge under mild conditions. Such an axiomatization is a fragment of the axiomatization for differential dynamic logic, and is therefore a finite extension of the axiomatization of real-closed fields. Furthermore, the numerical approximations approximate formulas containing special function symbols by $\text{FOL}_{\mathbb{R}}$ formulas, improving upon earlier decidability results only concerning closed sentences.

cs.LO↗
arXiv2025

Unfolding Boxes with Local Constraints

We consider the problem of finding and enumerating polyominos that can be folded into multiple non-isomorphic boxes. While several computational approaches have been proposed, including SAT, randomized algorithms, and decision diagrams, none has been able to perform at scale. We argue that existing SAT encodings are hindered by the presence of global constraints (e.g., graph connectivity or acyclicity), which are generally hard to encode effectively and hard for solvers to reason about. In this work, we propose a new SAT-based approach that replaces these global constraints with simple local constraints that have substantially better propagation properties. Our approach dramatically improves the scalability of both computing and enumerating common box unfoldings: (i) while previous approaches could only find common unfoldings of two boxes up to area 88, ours easily scales beyond 150, and (ii) while previous approaches were only able to enumerate common unfoldings up to area 30, ours scales up to 60. This allows us to rule out 46, 54, and 58 as the smallest areas allowing a common unfolding of three boxes, thereby refuting a conjecture of Xu et al. (2017).

cs.CG↗
arXiv2025

Automated Symmetric Constructions in Discrete Geometry

We present a computational methodology for obtaining rotationally symmetric sets of points satisfying discrete geometric constraints, and demonstrate its applicability by discovering new solutions to some well-known problems in combinatorial geometry. Our approach takes the usage of SAT solvers in discrete geometry further by directly embedding rotational symmetry into the combinatorial encoding of geometric configurations. Then, to realize concrete point sets corresponding to abstract designs provided by a SAT solver, we introduce a novel local-search realizability solver, which shows excellent practical performance despite the intrinsic $\exists \mathbb{R}$-completeness of the problem. Leveraging this combined approach, we provide symmetric extremal solutions to the Erd\H{o}s-Szekeres problem, as well as a minimal odd-sized solution with 21 points for the everywhere-unbalanced-points problem, improving on the previously known 23-point configuration. The imposed symmetries yield more aesthetically appealing solutions, enhancing human interpretability, and simultaneously offer computational benefits by significantly reducing the number of variables required to encode discrete geometric problems.

cs.DM↗
arXiv2025

MathPhys-Guided Coarse-to-Fine Anomaly Synthesis with SQE-Driven Bi-Level Optimization for Anomaly Detection

Currently, industrial anomaly detection suffers from two bottlenecks: (i) the rarity of real-world defect images and (ii) the opacity of sample quality when synthetic data are used. Existing synthetic strategies (e.g., cut-and-paste) overlook the underlying physical causes of defects, leading to inconsistent, low-fidelity anomalies that hamper model generalization to real-world complexities. In this paper, we introduce a novel and lightweight pipeline that generates synthetic anomalies through Math-Phys model guidance, refines them via a Coarse-to-Fine approach and employs a bi-level optimization strategy with a Synthesis Quality Estimator (SQE). By combining physical modeling of the three most typical physics-driven defect mechanisms: Fracture Line (FL), Pitting Loss (PL), and Plastic Warpage (PW), our method produces realistic defect masks, which are subsequently enhanced in two phases. The first stage (npcF) enforces a PDE-based consistency to achieve a globally coherent anomaly structure, while the second stage (npcF++) further improves local fidelity. Additionally, we leverage SQE-driven weighting, ensuring that high-quality synthetic samples receive greater emphasis during training. To validate our method, we conduct experiments on three anomaly detection benchmarks: MVTec AD, VisA, and BTAD. Across these datasets, our method achieves state-of-the-art results in both image- and pixel-AUROC, confirming the effectiveness of our MaPhC2F dataset and BiSQAD method. All code will be released.

cs.CV↗
arXiv2025

Bootstrapped Model Predictive Control

Model Predictive Control (MPC) has been demonstrated to be effective in continuous control tasks. When a world model and a value function are available, planning a sequence of actions ahead of time leads to a better policy. Existing methods typically obtain the value function and the corresponding policy in a model-free manner. However, we find that such an approach struggles with complex tasks, resulting in poor policy learning and inaccurate value estimation. To address this problem, we leverage the strengths of MPC itself. In this work, we introduce Bootstrapped Model Predictive Control (BMPC), a novel algorithm that performs policy learning in a bootstrapped manner. BMPC learns a network policy by imitating an MPC expert, and in turn, uses this policy to guide the MPC process. Combined with model-based TD-learning, our policy learning yields better value estimation and further boosts the efficiency of MPC. We also introduce a lazy reanalyze mechanism, which enables computationally efficient imitation learning. Our method achieves superior performance over prior works on diverse continuous control tasks. In particular, on challenging high-dimensional locomotion tasks, BMPC significantly improves data efficiency while also enhancing asymptotic performance and training stability, with comparable training time and smaller network sizes. Code is available at https://github.com/wertyuilife2/bmpc.

cs.LG↗
arXiv2025

Topolectrical space-time circuits

Topolectrical circuits have emerged as a pivotal platform for realizing static topological states that are challenging to construct in other systems, facilitating the design of robust circuit devices. In addition to spatial dimensionality, synergistic engineering of both temporal and spatial degrees in circuit networks holds tremendous potential across diverse technologies, such as wireless communications, non-reciprocal electronics and dynamic signal controls with exotic space-time topology. However, the realization of space-time modulated circuit networks is still lacking due to the necessity for flexible modulation of node connections in both spatial and temporal domains. Here, we propose a new class of topolectrical circuits, referred to as topolectrical space-time circuits, to bridge this gap. By designing and applying a novel time-varying circuit element controlled by external voltages, we can construct circuit networks exhibiting discrete space-time translational symmetries in any dimensionality, where the circuit dynamical equation is in the same form with time-dependent Schrodinger equation. Through the implementation of topolectrical space-time circuits, three distinct types of topological space-time crystals are experimentally demonstrated, including the (1+1)-dimensional topological space-time crystal with midgap edge modes, (2+1)-dimensional topological space-time crystal with chiral edge states, and (3+1)-dimensional Weyl space-time semimetals. Our work establishes a solid foundation for the exploration of intricate space-time topological phenomena and holds potential applications in the field of dynamically manipulating electronic signals with unique space-time topology

cond-mat.mes-hall↗
arXiv2024

What Matters in Building Vision-Language-Action Models for Generalist Robots

To utilize Foundation Vision Language Models (VLMs) for robotic tasks and motion planning, the community has proposed different methods for injecting action components into VLMs and building the Vision-Language-Action models (VLAs). In this work, we disclose the key factors that significantly influence the performance of VLA on robot manipulation problems and focus on answering three essential design choices: which backbone to select, how to formulate the VLA architectures, and when to add cross-embodiment data. The obtained results convince us firmly to explain why we prefer VLA and develop a new family of VLAs, RoboVLMs, which require very few manual designs and achieve a new state-of-the-art performance in three simulation tasks and real-world experiments. Through our extensive experiments, which include over 8 VLM backbones, 4 policy architectures, and over 600 distinct designed experiments, we provide a detailed guidebook for the future design of VLAs. In addition to the study, the highly flexible RoboVLMs framework, which supports easy integrations of new VLMs and free combinations of various design choices, is made public to facilitate future research. We open-source all details, including codes, models, datasets, and toolkits, along with detailed training and evaluation recipes at: robovlms.github.io.

cs.RO↗
arXiv2024

Friend or Foe? Harnessing Controllable Overfitting for Anomaly Detection

Overfitting has traditionally been viewed as detrimental to anomaly detection, where excessive generalization often limits models' sensitivity to subtle anomalies. Our work challenges this conventional view by introducing Controllable Overfitting-based Anomaly Detection (COAD), a novel framework that strategically leverages overfitting to enhance anomaly discrimination capabilities. We propose the Aberrance Retention Quotient (ARQ), a novel metric that systematically quantifies the extent of overfitting, enabling the identification of an optimal golden overfitting interval wherein model sensitivity to anomalies is maximized without sacrificing generalization. To comprehensively capture how overfitting affects detection performance, we further propose the Relative Anomaly Distribution Index (RADI), a metric superior to traditional AUROC by explicitly modeling the separation between normal and anomalous score distributions. Theoretically, RADI leverages ARQ to track and evaluate how overfitting impacts anomaly detection, offering an integrated approach to understanding the relationship between overfitting dynamics and model efficacy. We also rigorously validate the statistical efficacy of Gaussian noise as pseudo-anomaly generators, reinforcing the method's broad applicability. Empirical evaluations demonstrate that our controllable overfitting method achieves State-Of-The-Art(SOTA) performance in both one-class and multi-class anomaly detection tasks, thus redefining overfitting as a powerful strategy rather than a limitation.

cs.LG↗