arXiv subjects

Bowen Yang

Publications and source records attributed to Bowen Yang.

At least 55 records · Page 3Linked to original sources

arXiv2026

ComHymba: Low-Complexity Domain-Informed Foundation Model for Wireless Communications

Wireless foundation models are a promising route to unify channel reconstruction, sensing, and beam management in future wireless communication systems, but existing designs often inherit LLM-style Transformers with quadratic token complexity and weak integration of propagation priors. This paper proposes ComHymba, a domain-informed wireless foundation model built on an asymmetric masked autoencoder for large-scale self-supervised pre-training on Channel State Information (CSI). ComHymba introduces (i) 3D spatio-temporal-frequency patchification with rotary positional embedding, (ii) domain-informed masking strategies that emulate realistic CSI sparsity and fading patterns, and (iii) a decoupled amplitude--phase weighted objective tailored to channel statistics. Architecturally, we employ Hymba blocks that fuse windowed self-attention with state space models (SSMs), enabling linear-time modeling with respect to the overall channel input size. Experiments on eight downstream tasks spanning channel state information reconstruction, environmental sensing, and beam management show consistent accuracy gains over strong task-specific baselines, together with up to a $3.3\times$ inference speedup versus Transformer backbones. Overall, ComHymba provides a scalable and efficient backbone for AI-native physical-layer intelligence.

eess.SP↗

arXiv2026

Transition between 2D Symmetry Protected Topological Phases on a Klein Bottle

Manifolds with nontrivial topology play an essential role in the study of topological phases of matter. In this paper, we study the nontrivial symmetry response of the $2+1$D $Z_2$ symmetry-protected topological (SPT) phase when the system is put on a non-orientable manifold -- the Klein bottle. In particular, we find that when a symmetry defect is inserted along the orientation-reversing cycle of the Klein bottle, the ground state of the system gets an extra charge. This response remains well defined at transition points into the trivial SPT phase, resulting in an exact two-fold degeneracy in the ground state independent of the system size. We demonstrate the symmetry response using exactly solvable lattice models of the SPT phase, as well as numerical work across the transition. We explore the connection of this result to the modular transformation of the $3+1$D $Z_2$ gauge theory and the emergent nature of the parity symmetry in the $Z_2$ SPT phase.

cond-mat.str-el↗

arXiv2026

UnfoldLDM: Degradation-Aware Unfolding with Iterative Latent Diffusion Priors for Blind Image Restoration

Deep unfolding networks (DUNs) combine the interpretability of model-based methods with the learning ability of deep networks, yet remain limited for blind image restoration (BIR). Existing DUNs suffer from: (1) \textbf{Degradation-specific dependency}, as their optimization frameworks are tied to a known degradation model, making them unsuitable for BIR tasks; and (2) \textbf{Over-smoothing bias}, resulting from the direct feeding of gradient descent outputs, dominated by low-frequency content, into the proximal term, suppressing fine textures. To overcome these issues, we propose UnfoldLDM to integrate DUNs with latent diffusion model (LDM) for BIR. In each stage, UnfoldLDM employs a multi-granularity degradation-aware (MGDA) module as the gradient descent step. MGDA models BIR as an unknown degradation estimation problem and estimates both the holistic degradation matrix and its decomposed forms, enabling robust degradation removal. For the proximal step, we design a degradation-resistant LDM (DR-LDM) to extract compact degradation-invariant priors from the MGDA output. Guided by this prior, an over-smoothing correction transformer (OCFormer) explicitly recovers high-frequency components and enhances texture details. This unique combination ensures the final result is degradation-free and visually rich. Experiments show that our UnfoldLDM achieves a leading place on various BIR tasks and benefits downstream tasks. Moreover, our design is compatible with existing DUN-based methods, serving as a plug-and-play framework. Code will be released.

cs.CV↗

arXiv2026

Operator algebra and algorithmic construction of boundaries and defects in (2+1)D topological Pauli stabilizer codes

Quantum low-density parity-check codes, such as the Kitaev toric code and bivariate bicycle codes, are often defined with periodic boundary conditions, which are difficult to realize in physical systems. In this paper, we present an algorithm for constructing all gapped boundaries and defects of two-dimensional Pauli stabilizer codes. Using the operator algebra formalism, we establish a one-to-one correspondence between the topological data, such as anyon fusion rules and topological spins, of two-dimensional bulk stabilizer codes and one-dimensional boundary anomalous subsystem codes. To make the operator algebra computationally accessible, we adapt Laurent polynomials and convert the tasks into matrix operations, e.g., the Hermite normal form for obtaining boundary anyons and the Smith normal form for determining fusion rules. This approach enables computers to automatically generate all possible gapped boundaries and defects for topological Pauli stabilizer codes through boundary anyon condensation and topological order completion. This streamlines the analysis of surface codes and associated logical operations for fault-tolerant quantum computation. Our algorithm applies to $\mathbb{Z}_d$ qudits for both prime and nonprime $d$, enabling exploration of topological phases beyond the Kitaev toric code. We have applied the algorithm and explicitly demonstrated the lattice constructions of 2 boundaries and 6 defects in the $\mathbb{Z}_2$ toric code, 3 boundaries and 22 defects in the $\mathbb{Z}_4$ toric code, 1 boundary and 2 defects in the double semion code, 1 boundary and 22 defects in the six-semion code, 6 boundaries and 270 defects in the color code, and 6 defects in the anomalous three-fermion code. Finally, we study the boundaries of bivariate bicycle codes, showing that they exhibit large logical dimensions and anyons with long translation periods.

quant-ph↗

arXiv2026

Resolving spurious topological entanglement entropy in stabilizer codes

Topological entanglement entropy (TEE) is a key diagnostic of long-range entanglement in two-dimensional gapped phases of matter, but it can suffer from spurious contributions that overestimate the total quantum dimension of the underlying topological order. In this work, we identify the microscopic origin of spurious TEE and introduce a concave partition for computing the Levin-Wen TEE of translation-invariant stabilizer codes of prime-dimensional qudits. We rigorously prove that this prescription is free of spurious contributions. As a complementary probe, we study bivariate bicycle codes on a bipartite cylinder and show that the entanglement entropy depends sensitively on the cylinder circumference, revealing topological frustration of the underlying anyons.

quant-ph↗

arXiv2026

The Classification of Pauli Stabilizer Codes: A Lattice and Continuum Treatise

We classify mobile Pauli stabilizer codes up to gapped interfaces and coarse-graining using the framework of algebraic $\mathrm{L}$-theory. We compare this classification with that of framed TQFTs, theories that arise naturally in the continuum, highlighting a close structural relationship between the two. Our approach is formulated in the category of perfect chain complexes equipped with quadratic functor over the Laurent polynomial ring $R = \mathbb{Z}/p[x_1^{\pm 1}, \ldots, x_n^{\pm 1}]$, within which the collection of topological operators of Pauli stabilizer codes arise naturally as objects. In particular, we establish a bulk-boundary correspondence for lattice theories: the equivalence class of a Pauli stabilizer code up to gapped interface is described by a Clifford QCA in one dimension higher. This is done using the universal target category for stabilizer codes, which is the categorical spectrum whose existence and universal properties are introduced in this work. We conclude by highlighting subtle differences between the classification of Pauli stabilizer codes and TQFTs, leading to qualitative distinctions between lattice and continuum theories.

math-ph↗

arXiv2026

RoboCOIN: An Open-Sourced Bimanual Robotic Data Collection for Integrated Manipulation

Despite the critical role of bimanual manipulation in endowing robots with human-like dexterity, large-scale and diverse datasets remain scarce due to the significant hardware heterogeneity across bimanual robotic platforms. To bridge this gap, we introduce RoboCOIN, a large-scale multi-embodiment bimanual manipulation dataset comprising over 180,000 demonstrations collected from 15 distinct robotic platforms. Spanning 16 diverse environments-including residential, commercial, and industrial settings-the dataset features 421 bimanual tasks systematically categorized by 39 bimanual collaboration actions and 432 objects. A key innovation of our work is the hierarchical capability pyramid, which provides granular annotations ranging from trajectory-level concepts to segment-level subtasks and frame-level kinematics. Furthermore, we present CoRobot, an efficient data processing pipeline powered by the Robot Trajectory Markup Language (RTML), designed to facilitate quality assessment, automated annotation, and unified multi-embodiment and data management. Extensive experiments demonstrate the effectiveness of RoboCOIN in enhancing the performance of various bimanual manipulation models across a wide spectrum of robotic embodiments. The entire dataset and codebase are fully open-sourced, providing a valuable resource for advancing research in bimanual and multi-embodiment manipulation.

cs.RO↗

arXiv2026

Voxtral TTS

We introduce Voxtral TTS, an expressive multilingual text-to-speech model that generates natural speech from as little as 3 seconds of reference audio. Voxtral TTS adopts a hybrid architecture that combines auto-regressive generation of semantic speech tokens with flow-matching for acoustic tokens. These tokens are encoded and decoded with Voxtral Codec, a speech tokenizer trained from scratch with a hybrid VQ-FSQ quantization scheme. In human evaluations conducted by native speakers, Voxtral TTS is preferred for multilingual voice cloning due to its naturalness and expressivity, achieving a 68.4\% win rate over ElevenLabs Flash v2.5. We release the model weights under a CC BY-NC license.

cs.AI↗

arXiv2026

Intern-S1-Pro: Scientific Multimodal Foundation Model at Trillion Scale

We introduce Intern-S1-Pro, the first one-trillion-parameter scientific multimodal foundation model. Scaling to this unprecedented size, the model delivers a comprehensive enhancement across both general and scientific domains. Beyond stronger reasoning and image-text understanding capabilities, its intelligence is augmented with advanced agent capabilities. Simultaneously, its scientific expertise has been vastly expanded to master over 100 specialized tasks across critical science fields, including chemistry, materials, life sciences, and earth sciences. Achieving this massive scale is made possible by the robust infrastructure support of XTuner and LMDeploy, which facilitates highly efficient Reinforcement Learning (RL) training at the 1-trillion parameter level while ensuring strict precision consistency between training and inference. By seamlessly integrating these advancements, Intern-S1-Pro further fortifies the fusion of general and specialized intelligence, working as a Specializable Generalist, demonstrating its position in the top tier of open-source models for general capabilities, while outperforming proprietary models in the depth of specialized scientific tasks.

cs.LG↗

arXiv2026

Discontinuous transition to shear flow turbulence

Depending on the type of flow, the transition to turbulence can take one of two forms: either turbulence arises from a sequence of instabilities or from the spatial proliferation of transiently chaotic domains, a process analogous to directed percolation. The former scenario is commonly referred to as a supercritical transition and frequently encountered in flows destabilized by body forces, whereas the latter subcritical transition is common in shear flows. Both cases are inherently continuous in a sense that the transformation from ordered laminar to fully turbulent fluid motion is only accomplished gradually with flow speed. Here we show that these established transition types do not account for the more general setting of shear flows subject to body forces. The combination of the two continuous scenarios leads to the attenuation of spatial coupling; with increasing forcing amplitude, the transition becomes increasingly sharp and eventually discontinuous. We argue that the suppression of laminar-turbulent coexistence and the approach towards a discontinuous phase transition potentially apply to a broad range of situations including flows subject to, for example, buoyancy, centrifugal or electromagnetic forces.

physics.flu-dyn↗

arXiv2026

Clifford quantum cellular automata from topological quantum field theories and invertible subalgebras

We present a general framework for constructing quantum cellular automata (QCA) from topological quantum field theories (TQFT) and invertible subalgebras (ISA) using the cup-product formalism. This approach explicitly realizes all $\mathbb{Z}_2$ and $\mathbb{Z}_p$ Clifford QCAs (for prime $p$) in all admissible dimensions, in precise agreement with the classification predicted by algebraic $L$-theory. We determine the orders of these QCAs by explicitly showing that finite powers reduce to the identity up to finite-depth quantum circuits (FDQC) and lattice translations. In particular, we demonstrate that the $\mathbb{Z}_2$ Clifford QCAs in $(4l{+}1)$ spatial dimensions can be disentangled by non-Clifford FDQCs. Our construction applies beyond cubic lattices, allowing $\mathbb{Z}_2$ QCAs to be defined on arbitrary cellulations. Furthermore, we explicitly construct invertible subalgebras in higher dimensions, obtaining $\mathbb{Z}_2$ ISAs in $2l$ spatial dimensions and $\mathbb{Z}_p$ ISAs in $(4l{-}2)$ spatial dimensions. These ISAs give rise to $\mathbb{Z}_2$ QCAs in $(2l{+}1)$ dimensions and $\mathbb{Z}_p$ QCAs in $(4l{-}1)$ dimensions. We further prove that the QCAs in $3$ spatial dimensions constructed via TQFTs and ISAs are equivalent by identifying their boundary algebras, and show that this approach extends to higher dimensions. Together, these results establish a unified and dimension-periodic framework for Clifford QCAs, connecting their explicit lattice realizations to field theories.

math.QA↗

arXiv2026

A mathematical theory of topological invariants of quantum lattice systems

We show that Hall conductance and its non-abelian and higher-dimensional analogs are obstructions to promoting a symmetry of a state to a gauge symmetry. To do this, we define a local Lie algebra over a Grothendieck site as a pre-cosheaf of Lie algebras with additional properties and propose that a gauge symmetry should be described by such an object. We show that infinitesimal symmetries of a gapped state of a quantum spin system form a local Lie algebra over a site of semilinear sets and use it to construct topological invariants of the state. Our construction applies to lattice systems on arbitrary asymptotically conical subsets of a Euclidean space including those which cannot be studied using field theory.

math-ph↗

arXiv2026

OS-Themis: A Scalable Critic Framework for Generalist GUI Rewards

Reinforcement Learning (RL) has the potential to improve the robustness of GUI agents in stochastic environments, yet training is highly sensitive to the quality of the reward function. Existing reward approaches struggle to achieve both scalability and performance. To address this, we propose OS-Themis, a scalable and accurate multi-agent critic framework. Unlike a single judge, OS-Themis decomposes trajectories into verifiable milestones to isolate critical evidence for decision making and employs a review mechanism to strictly audit the evidence chain before making the final verdict. To facilitate evaluation, we further introduce OmniGUIRewardBench (OGRBench), a holistic cross-platform benchmark for GUI outcome rewards, where all evaluated models achieve their best performance under OS-Themis. Extensive experiments on AndroidWorld show that OS-Themis yields a 10.3% improvement when used to support online RL training, and a 6.9% gain when used for trajectory validation and filtering in the self-training loop, highlighting its potential to drive agent evolution.

cs.AI↗

arXiv2026

AoE: Always-on Egocentric Human Video Collection for Embodied AI

Embodied foundation models require large-scale, high-quality real-world interaction data for pre-training and scaling. However, existing data collection methods suffer from high infrastructure costs, complex hardware dependencies, and limited interaction scope, making scalable expansion challenging. In fact, humans themselves are ideal physically embodied agents. Therefore, obtaining egocentric real-world interaction data from globally distributed "human agents" offers advantages of low cost and sustainability. To this end, we propose the Always-on Egocentric (AoE) data collection system, which aims to simplify hardware dependencies by leveraging humans themselves and their smartphones, enabling low-cost, highly efficient, and scene-agnostic real-world interaction data collection to address the challenge of data scarcity. Specifically, we first employ an ergonomic neck-mounted smartphone holder to enable low-barrier, large-scale egocentric data collection through a cloud-edge collaborative architecture. Second, we develop a cross-platform mobile APP that leverages on-device compute for real-time processing, while the cloud hosts automated labeling and filtering pipelines that transform raw videos into high-quality training data. Finally, the AoE system supports distributed Ego video data collection by anyone, anytime, and anywhere. We evaluate AoE on data preprocessing quality and downstream tasks, demonstrating that high-quality egocentric data significantly boosts real-world generalization.

cs.CV↗

arXiv2026

Construction-Verification: A Benchmark for Applied Mathematics in Lean 4

Recent advances in large language models have demonstrated impressive capabilities in mathematical formalization. However, existing benchmarks focus on logical verification of declarative propositions, often neglecting the task of explicitly synthesizing solutions. This limitation is particularly acute in applied mathematics domains, where the goal is frequently to derive concrete values or executable algorithms rather than solely proving theorems. To address this, we introduce a Lean 4 framework that enforces a construction-verification workflow, compelling the agent to define explicit solutions before proving their correctness. We curate a comprehensive benchmark AMBER (Applied Mathematics BEnchmark for Reasoning) spanning core domains of applied mathematics, including convex analysis, optimization, numerical algebra, and high-dimensional probability. Aside from theorem proving, our benchmark features complex tasks such as evaluation, algorithm design, and representation transformation. Experiments reveal that current models face significant difficulties with these constructive tasks. Notably, we observe that general-purpose reasoning models consistently outperform specialized theorem provers. We attribute this to a degradation of instruction following capabilities in specialized models. Fine-tuning on proof corpora appears to induce ``tactical overfitting", compromising the ability to adhere to complex constructive requirements, whereas general models retain the versatility needed for multi-task formal reasoning.

cs.LO↗

arXiv2026

OmniEVA: Embodied Versatile Planner via Task-Adaptive 3D-Grounded and Embodiment-aware Reasoning

Recent advances in multimodal large language models (MLLMs) have opened new opportunities for embodied intelligence, enabling multimodal understanding, reasoning, and interaction, as well as continuous spatial decision-making. Nevertheless, current MLLM-based embodied systems face two critical limitations. First, Geometric Adaptability Gap: models trained solely on 2D inputs or with hard-coded 3D geometry injection suffer from either insufficient spatial information or restricted 2D generalization, leading to poor adaptability across tasks with diverse spatial demands. Second, Embodiment Constraint Gap: prior work often neglects the physical constraints and capacities of real robots, resulting in task plans that are theoretically valid but practically infeasible. To address these gaps, we introduce OmniEVA -- an embodied versatile planner that enables advanced embodied reasoning and task planning through two pivotal innovations: (1) a Task-Adaptive 3D Grounding mechanism, which introduces a gated router to perform explicit selective regulation of 3D fusion based on contextual requirements, enabling context-aware 3D grounding for diverse embodied tasks. (2) an Embodiment-Aware Reasoning framework that jointly incorporates task goals and embodiment constraints into the reasoning loop, resulting in planning decisions that are both goal-directed and executable. Extensive experimental results demonstrate that OmniEVA not only achieves state-of-the-art general embodied reasoning performance, but also exhibits a strong ability across a wide range of downstream scenarios. Evaluations of a suite of proposed embodied benchmarks, including both primitive and composite tasks, confirm its robust and versatile planning capabilities. Project page: https://omnieva.github.io

cs.RO↗

arXiv2026

A Training-Free Guess What Vision Language Model from Snippets to Open-Vocabulary Object Detection

Open-Vocabulary Object Detection (OVOD) aims to develop the capability to detect anything. Although myriads of large-scale pre-training efforts have built versatile foundation models that exhibit impressive zero-shot capabilities to facilitate OVOD, the necessity of creating a universal understanding for any object cognition according to already pretrained foundation models is usually overlooked. Therefore, in this paper, a training-free Guess What Vision Language Model, called GW-VLM, is proposed to form a universal understanding paradigm based on our carefully designed Multi-Scale Visual Language Searching (MS-VLS) coupled with Contextual Concept Prompt (CCP) for OVOD. This approach can engage a pre-trained Vision Language Model (VLM) and a Large Language Model (LLM) in the game of "guess what". Wherein, MS-VLS leverages multi-scale visual-language soft-alignment for VLM to generate snippets from the results of class-agnostic object detection, while CCP can form the concept of flow referring to MS-VLS and then make LLM understand snippets for OVOD. Finally, the extensive experiments are carried out on natural and remote sensing datasets, including COCO val, Pascal VOC, DIOR, and NWPU-10, and the results indicate that our proposed GW-VLM can achieve superior OVOD performance compared to the-state-of-the-art methods without any training step.

cs.CV↗

arXiv2026

OS-Symphony: A Holistic Framework for Robust and Generalist Computer-Using Agent

While Vision-Language Models (VLMs) have significantly advanced Computer-Using Agents (CUAs), current frameworks struggle with robustness in long-horizon workflows and generalization in novel domains. These limitations stem from a lack of granular control over historical visual context curation and the absence of visual-aware tutorial retrieval. To bridge these gaps, we introduce OS-Symphony, a holistic framework that comprises an Orchestrator coordinating two key innovations for robust automation: (1) a Reflection-Memory Agent that utilizes milestone-driven long-term memory to enable trajectory-level self-correction, effectively mitigating visual context loss in long-horizon tasks; (2) Versatile Tool Agents featuring a Multimodal Searcher that adopts a SeeAct paradigm to navigate a browser-based sandbox to synthesize live, visually aligned tutorials, thereby resolving fidelity issues in unseen scenarios. Experimental results demonstrate that OS-Symphony delivers substantial performance gains across varying model scales, establishing new state-of-the-art results on three online benchmarks, notably achieving 65.84% on OSWorld.

cs.MA↗