arXiv subjects

Rui Chen

Publications and source records attributed to Rui Chen.

At least 91 records · Page 5Linked to original sources

arXiv2026

A PEMFC-based combined cooling heating and power (CCHP) system with flexible energy supply: Thermodynamic and economic analyses

Proton exchange membrane fuel cell (PEMFC) systems offer a key approach to hydrogen utilization, and PEMFC-based combined cooling, heating, and power (CCHP) systems pave the way for an efficient and clean energy supply to buildings. In conventional PEMFC-CCHP systems, the heating/cooling capacity and electrical power output are strongly coupled, making it difficult to meet diverse energy demands. This paper presents a novel energy system that integrates an organic Rankine cycle and an absorption heat pump in a parallel configuration, which enables flexible regulation of electricity-cooling capacities in summer and electricity-heating capacities in winter by adjusting the splitting ratio of the waste heat. The impacts of the splitting ratio and key operating parameters on thermodynamic performance and economic performance are quantitatively evaluated. The results show that the ORC can improve electrical efficiency by 2.19 percentage points in summer and 2.78 percentage points in winter. When the current density is fixed at 0.4 A/cm2 and the splitting ratio increases from 0 to 0.5, the cooling capacity of the system varies from 1294 to 647 W, and the heating capacity varies from 2660 to 1330 W. The economic performance is more sensitive to electricity price and hydrogen price than to other parameters, confirmed by their high sensitivity coefficients for net present value (NPV) and internal rate of return (IRR). This system possesses excellent thermodynamic and economic properties, thereby offering significant potential for reducing building energy consumption and carbon emissions.

physics.chem-ph↗

arXiv2026

A Residual-Subspace Constraint Framework for Fourier Ptychographic Microscopy

The reconstruction fidelity of computational optical imaging is fundamentally constrained by the model-reality gap, i.e., the inevitable discrepancy between idealized forward models and the physical imaging process. Conventional paradigms attempt to bridge this gap through exhaustive system calibration or explicit parameter estimation, which are often computationally intensive and prone to severe non-convex stagnation. This paper introduces a Residual-Subspace Constraint Framework (RSCF) to achieve robust Fourier ptychographic microscopy. Instead of treating residuals as unstructured errors, RSCF leverages subspace decomposition to decouple low-rank, systematic mismatches from stochastic noise, thereby isolating stable information manifolds that remain invariant to forward-model inaccuracies. By embedding this subspace constraint into the iterative engine, the framework selectively suppresses error-amplifying components, enabling high-fidelity phase and amplitude recovery without explicit hardware calibration. Numerical simulations and experimental validations demonstrate that RSCF yields superior convergence acceleration and artifact suppression under severe optical aberrations and LED misalignment. This information-centric paradigm provides a versatile, model-agnostic strategy to enhance robustness across diverse computational imaging modalities.

physics.optics↗

arXiv2026

What if Agents Could Imagine? Reinforcing Open-Vocabulary HOI Comprehension through Generation

Multimodal Large Language Models have shown promising capabilities in bridging visual and textual reasoning, yet their reasoning capabilities in Open-Vocabulary Human-Object Interaction (OV-HOI) are limited by cross-modal hallucinations and limited viewpoints of images. To address this, we propose ImagineAgent, an agentic framework that integrates cognitive mapping, tool-augmented reinforcement learning (RL), and generative world modeling for robust OV-HOI understanding. Specifically, we first propose an innovative CoT dataset named hicodet-6K for supervised fine-tuning (SFT), which effectively bridges the perception-to-cognition gap by structuring perceived entities into interaction pairs for comprehensive predictions. Subsequently, we develop a multimodal tool library integrating online retrieval, image cropping, and generative modeling, enabling the agent to dynamically augment reasoning with domain-specific tools to resolve visual-semantic ambiguities and hallucinations during inference. Moreover, we incorporate a generative model to reconstruct alternative viewpoints, enabling the agent to 'imagine' under limited viewpoints. Finally, we propose a composite reward mechanism to jointly optimize prediction accuracy and tool efficiency. Evaluations on both SWIG-HOI and HICO-DET datasets demonstrate that our method achieves state-of-the-art performance while requiring merely 36.7% of the training data compared to existing methods, validating our robustness, empirical effectiveness and efficiency.

cs.CV↗

arXiv2026

SAVER: Selective As-Needed Vision Evidence for Multimodal Information Extraction

Multimodal IE in social media is difficult because a post may attach multiple images that are weakly related, redundant, or even misleading with respect to the text. In this setting, always-on multimodal fusion wastes computation and can amplify spurious visual cues. The core challenge is to decide, for each candidate span or marked entity pair, whether vision should be consulted at all and, if so, which small subset of images provides trustworthy evidence. We propose SAVER, a selective vision-as-needed framework for multimodal named entity recognition and multimodal relation extraction. SAVER uses a Conformal Groundability Gate (CGG) to estimate span-level visual groundability in MNER, derive pair-level activation in MRE from the two marked entities, and calibrate the activation threshold on a held-out split via a conformal-style procedure with Clopper--Pearson upper bounds. When activated, a submodular relevance--diversity selector chooses a compact evidence subset across images, which is then aggregated by a Set Transformer. An energy-inspired joint scoring head combines text, optional visual evidence, text--image consistency, and sparse routing for entity typing or relation classification. Experiments show that SAVER consistently improves F1 over strong text-only and always-on multimodal baselines, while reducing AURC, increasing activation coverage at a fixed risk level, and lowering FLOPs and P90 latency.

cs.CV↗

arXiv2026

SCOPE and SCION: A Benchmark and an Auditable Reference Pipeline for Schema Induction and Fusion from Text

Schema graphs are an upstream bottleneck of schema-grounded information extraction and knowledge graph construction, yet most extraction systems assume the schema is already available. We introduce SCOPE (Schema Construction and Ontology-induction Pipeline Evaluation), a train-text-only benchmark for corpus-to-schema induction and optional schema fusion from raw text, built from 24 public information extraction sources (15 RE and 9 EE) normalized into evaluation-only gold schema graphs; its core event-extraction target covers event types and within-event argument roles, with inter-event links reported separately. We present SCION (Schema Construction and Induction with Ontology Normalization), an auditable reference pipeline rather than a new extraction architecture; it constructs candidate spaces from train text and restricts naming, merging, filtering, validation, and conservative fusion to candidate-linked evidence under strict JSON contracts. On the SCOPE core suite, SCION-lite attains the highest F1 among released source-schema references, Text2Onto-style, LLM-only, and matched extract-then-aggregate baselines under Literal, Fuzzy, Continuous, and Graph schema-graph metrics, while the compact open-model SCION-RL variant reduces reliance on proprietary LLM schema engineers. These results are reported against normalized typed-edge targets rather than as claims that induced schemas surpass human ontology design; the release includes evidence-linked outputs, parse/fallback logs, candidate retention/merging logs, run manifests, code, and benchmark packages at https://github.com/wandugu/paper_scion.

cs.AI↗

arXiv2026

Dexora: Open-source VLA for High-DoF Bimanual Dexterity

Vision-Language-Action (VLA) models have recently become a central direction in embodied AI, but current systems are restricted to either dual-gripper control or single-arm dexterous hand manipulation. While low-dimensional gripper control can often be handled with simpler methods, high-dimensional dexterous hand control benefits greatly from full end-to-end VLA learning. In this work, we introduce Dexora, the first open-source VLA system that natively targets dual-arm, dual-hand high-DoF manipulation. We design a hybrid teleoperation pipeline that decouples gross arm kinematics (captured with a custom exoskeleton backpack) from fine finger motion (markerless hand tracking via Apple Vision Pro), and that drives both a physical dual-arm dual-hand platform and an identical MuJoCo digital twin. Using that interface, we assemble a large training corpus: an embodiment-matched synthetic corpus (100K simulated trajectories, 6.5M frames) and a real-world dataset of 10K teleoperated episodes (2.92M frames). To mitigate noisy teleoperation demonstrations, we propose a data-quality-aware training recipe: an offline discriminator provides clip-level weights for diffusion-transformer policy training, down-weighting low-quality demonstrations. Empirically, Dexora outperforms competitive VLA baselines on both basic and dexterous benchmarks (e.g., average dexterous success 66.7% vs. 51.7%), attains 90% success on basic tasks, and shows robust out-of-distribution and cross-embodiment generalization. Ablations confirm the importance of real data and the discriminator for dexterity.

cs.RO↗

arXiv2026

Effect of startup modes on cold start performance of PEM fuel cells with different cathode flow fields

Proton Exchange Membrane Fuel Cell (PEMFC) is widely recognized for its cleanliness and high efficiency, but is still facing challenges in cold environments. At low temperatures, the formation of ice and repeated freezing/thawing cycles may cause cell performance reduction and irreversible degradation. The cathode flow field of PEMFCs has a significant effect on the performance. In contrast to the conventional ``channel-ridge'' flow field, the metal foam has the advantages of excellent pre-distribution of gases and water drainage, which make it a promising candidate for the cold start. This paper examines the cold start of PEMFCs with metal foam flow field (MFFF) and serpentine flow field (SFF), and the influence of constant current mode, constant voltage mode, and ramping current mode is investigated experimentally through performance test and electrochemical characterization. The results show that lowering the voltage and increasing the current can enhance the cold-start performance of fuel cells. The MFFF fuel cell has superior cold start performance compared to the SFF fuel cell under the constant voltage mode of 0.3 V. Furthermore, the variable current mode is developed by considering the distinct properties of heat and water production during various phases, and the results indicate that increasing the current density at the unsaturated stage leads to an elevated rate of heat production and a reduced rate of water production, which can improve the cold start of PEMFCs.

physics.flu-dyn↗

arXiv2026

Open-Source Image Editing Models Are Zero-Shot Vision Learners

Recent studies have shown that large generative models can solve vision tasks they were not explicitly trained for. However, existing evidence relies on closed-source models~(Veo~3, Nano Banana Pro) or requires task-specific instruction tuning, leaving open whether publicly available image-editing models possess zero-shot vision abilities out of the box. We conduct a systematic evaluation of three open-source image-editing models -- Qwen-Image-Edit, FireRed-Image-Edit, and LongCat-Image-Edit -- on dense visual prediction tasks \emph{without any fine-tuning}. We benchmark monocular depth estimation on NYUv2 and DIODE, surface normal estimation on NYUv2, and semantic segmentation on Cityscapes, covering both geometric and semantic scene understanding. Results show that open-source image-editing models exhibit non-trivial zero-shot visual understanding. On NYUv2 surface normals, FireRed-Image-Edit achieves a mean angular error of $17.69^\circ$, surpassing the fine-tuned Marigold ($20.86^\circ$) and matching the instruction-tuned Vision Banana ($17.78^\circ$) without any task-specific training. On NYUv2 depth estimation, LongCat-Image-Edit obtains $δ_1{=}0.822$ with affine alignment, and Qwen-Image-Edit leads on DIODE Indoor ($δ_1{=}0.868$). On Cityscapes semantic segmentation, Qwen-Image-Edit reaches 25.7 mIoU at the 19-class level and 49.5 mIoU at a coarser 7-category level. By comparing three independently trained editors, we test whether zero-shot vision ability is an emergent property of image-editing pretraining rather than a model-specific artifact. Code, evaluation scripts, and all results are publicly released to serve as a reproducible baseline for future work.

cs.CV↗

arXiv2026

Tunable cornerlike states in topological type-II hyperbolic lattices

Type-II hyperbolic lattices constitute a new class of hyperbolic structures that are projected onto the Poincaré ring and possess both an inner and an outer boundary. In this work, we reveal the higher-order topological phases in type-II hyperbolic lattices, characterized by the generalized quadrupole moment. Unlike the type-I hyperbolic lattices where zero-energy cornerlike states exist on a single boundary, the higher-order topological phases in type-II hyperbolic lattices possess zero-energy cornerlike states localized on both the inner and outer boundaries. These findings are verified within both the modified Bernevig-Hughes-Zhang model and the Benalcazar-Bernevig-Hughes model. Furthermore, we demonstrate that the higher-order topological phase remains robust against weak disorder in type-II hyperbolic lattices. Our work provides a route for realizing and controlling higher-order topological states in type-II hyperbolic lattices.

cond-mat.mes-hall↗

arXiv2026

OmniAlpha: Aligning Transparency-Aware Generation via Multi-Task Unified Reinforcement Learning

Transparency-aware generation requires modeling not only RGB appearance but also alpha-based opacity and cross-layer composition, which are essential for tasks such as image matting, object removal, layer decomposition, and multi-layer content creation. However, existing RGBA-related methods remain largely fragmented, with separate pipelines designed for individual tasks. While a unified model is desirable, supervised fine-tuning alone is insufficient, as localized regression objectives cannot directly optimize the compositional fidelity, alpha-boundary precision, and structural consistency required for high-quality RGBA generation. To address this, we propose OmniAlpha, a unified multi-task reinforcement learning framework for transparency-aware generation and manipulation. OmniAlpha combines an end-to-end alpha-aware VAE and a sequence-to-sequence Diffusion Transformer, with a bi-directional layer axis in positional encoding to jointly model multiple RGBA inputs and outputs within a single forward pass. Built on a multi-task SFT cold start, it further performs GRPO-style post-training with layer-aware rewards defined on decoded RGBA outputs, enabling direct optimization of cross-layer coherence and fine transparency details. Experiments across five categories of transparency-aware tasks show that OmniAlpha consistently outperforms its unified SFT baseline and achieves strong performance against specialized expert models, including a 9.07% relative reduction in RGB L1 on layer decomposition and 74%/68% improvements over conventional matting tools on SAD/Grad for automatic matting.

cs.CV↗

arXiv2026

Enforcing TSP-Optimality in Fair Vehicle Routing by Cutting Planes

We study the fair capacitated vehicle routing problem, in which a fleet of vehicles must serve a set of customers such that the difference between the longest and shortest route, the range, is minimized. A key challenge is that the range objective is non-monotonic: it can be reduced by artificially lengthening routes, leading to solutions that violate TSP-optimality of individual routes. Existing exact methods struggle to handle this efficiently. We propose a branch-price-and-cut framework that enforces TSP-optimality through TSP-optimality cuts, which forbid TSP-dominated arc sequences. We strengthen the cuts through a dedicated lifting procedure. Computational experiments on benchmark instances with up to 25 customers show the method solves nearly all instances to optimality, achieving an average gap of 0.27% on the hardest configurations.

math.OC↗

arXiv2026

Seed3D 2.0: Advancing High-Fidelity Simulation-Ready 3D Content Generation

We present Seed3D 2.0, an advanced 3D content generation system built on Seed3D 1.0, with substantial improvements across generation fidelity, simulation-ready capabilities, and application coverage. For geometry, a coarse-to-fine two-stage pipeline decouples global structure learning from high-frequency detail recovery, while a locality-aware VAE achieves higher spatial compression and more efficient decoding. For texture and material generation, we replace the cascaded pipeline of Seed3D 1.0 with a unified PBR model that directly generates multi-view albedo and metallic-roughness maps, enhanced by Mixture-of-Experts scaling and VLM-based semantic conditioning for improved material precision and visual fidelity. Beyond single-object generation, Seed3D 2.0 introduces a simulation-ready model suite comprising scene layout planning, part-aware decomposition, and training-free articulation generation, enabling coherent scene construction and part-level physical interaction across physics and graphics engines. A large-scale human preference study against five recent commercial models shows that Seed3D 2.0 achieves consistent win rates of 69.0% to 89.9% in textured 3D asset generation. Seed3D 2.0 is available on https://exp.volcengine.com/ark/vision?_vtm_=0.0.c70961.d701978.0&mode=vision&modelId=doubao-seed3d-2-0-260328&tab=Gen3D

cs.GR↗

arXiv2026

Automated LTL Specification Generation from Industrial Aerospace Requirements

In the development and verification of safety-critical aero-space software, Linear Temporal Logic (LTL) has been widely used to specify complex system properties derived from requirements. However, a significant gap remains in industrial practice: translating natural language (NL) requirements into formal LTL properties is a labor-intensive and error-prone process that requires rare expertise in both aerospace control engineering and formal methods. While recent NL-to-LTL tools (e.g., NL2SPEC, NL2TL, NL2LTL) are capable of automating parts of this process, they often fail on real requirement documents in industrial settings, due to complex domain terminology or implicit temporal and logical structure. To address these challenges, we present AeroReq2LTL, a framework that automates LTL property generation for aerospace requirements using large language models (LLMs), with two key industrial innovations: (i) a data dictionary that normalizes technical jargon into precise atomic propositions; and (ii) a template-based requirement language that makes temporal cues and logical relations explicit before translation. On a real aerospace dataset, AeroReq2LTL achieves 85% precision and 88% recall in LTL generation, and its outputs can be directly consumed by existing verification tools.

cs.SE↗

arXiv2026

Layer Hall effect induced by altermagnetism

In this work, we propose a scheme to realize the layer Hall effect in the ferromagnetic topological insulator Bi$_2$Se$_3$ via proximity to $d$-wave altermagnets. We show that an altermagnet and an in-plane magnetic field applied near one surface gap the corresponding Dirac cone, yielding an altermagnet-induced half-quantized Hall effect. When altermagnets with antiparallel Néel vectors are placed near the top and bottom surfaces, giving rise to the layer Hall effect with vanishing net Hall conductance, i.e., the altermagnet-induced layer Hall effect. In contrast, altermagnets with parallel Néel vectors lead to a quantized Chern insulating state, i.e., the altermagnet-induced anomalous Hall effect. We further analyze the dependence of the Hall conductance on the orientation of the in-plane magnetic field and demonstrate that the layer Hall effect becomes observable under a perpendicular electric field. Our results establish a route to engineer altermagnet-induced topological phases in ferromagnetic topological insulators.

cond-mat.mes-hall↗

arXiv2026

AutoDrive-R$^2$: Incentivizing Reasoning and Self-Reflection Capacity for VLA Model in Autonomous Driving

Vision-Language-Action (VLA) models in autonomous driving systems have recently demonstrated transformative potential by integrating multimodal perception with decision-making capabilities. However, the interpretability and coherence of the decision process and the plausibility of action sequences remain largely underexplored. To address these issues, we propose AutoDrive-R$^2$, a novel VLA framework that enhances both reasoning and self-reflection capabilities of autonomous driving systems through chain-of-thought (CoT) processing and reinforcement learning (RL). Specifically, we first propose an innovative CoT dataset named nuScenesR$^2$-6K for supervised fine-tuning, which effectively builds cognitive bridges between input information and output trajectories through a four-step logical chain with self-reflection for validation. Moreover, to maximize both reasoning and self-reflection during the RL stage, we further employ the Group Relative Policy Optimization (GRPO) algorithm within a physics-grounded reward framework that incorporates spatial alignment, vehicle dynamic, and temporal smoothness criteria to ensure reliable and realistic trajectory planning. Extensive evaluation results across both nuScenes and Waymo datasets demonstrates the state-of-the-art performance and robust generalization capacity of our proposed method.

cs.RO↗

arXiv2026

Video-STAR: Reinforcing Open-Vocabulary Action Recognition with Tools

Multimodal large language models (MLLMs) have demonstrated remarkable potential in bridging visual and textual reasoning, yet their reliance on text-centric priors often limits their ability to disentangle semantically similar actions in open-vocabulary scenarios. To address this, we propose Video-STAR, a framework that harmonizes contextual sub-motion decomposition with tool-augmented reinforcement learning for open-vocabulary action recognition (OVAR). Unlike prior methods that treat actions as monolithic entities, our approach innovatively decomposes actions into discriminative sub-motions for fine-grained matching while dynamically invoking domain-specific tools for cross-modal interleaving, thereby enabling category-specific reasoning capacity and reducing cross-modal hallucination. Moreover, by designing a hierarchical reward that balances tool-usage efficiency, sub-motion relevance, and structural coherence in reasoning, our method autonomously leverages external tools to prioritize sub-motion patterns without explicit supervision, transmitting from text-centric reasoning to visually grounded inference. Extensive evaluations on HMDB-51, UCF-101, SSv2, Kinetics-400, and Kinetics-600 datasets demonstrate our state-of-the-art performance, outperforming existing methods in distinguishing fine-grained actions and handling cross-modal hallucination, validating our excellent robustness and generalization.

cs.CV↗

arXiv2026

HY-World 2.0: A Multi-Modal World Model for Reconstructing, Generating, and Simulating 3D Worlds

We introduce HY-World 2.0, a multi-modal world model framework that advances our prior project HY-World 1.0. HY-World 2.0 accommodates diverse input modalities, including text prompts, single-view images, multi-view images, and videos, and produces 3D world representations. With text or single-view image inputs, the model performs world generation, synthesizing high-fidelity, navigable 3D Gaussian Splatting (3DGS) scenes. This is achieved through a four-stage method: a) Panorama Generation with HY-Pano 2.0, b) Trajectory Planning with WorldNav, c) World Expansion with WorldStereo 2.0, and d) World Composition with WorldMirror 2.0. Specifically, we introduce key innovations to enhance panorama fidelity, enable 3D scene understanding and planning, and upgrade WorldStereo, our keyframe-based view generation model with consistent memory. We also upgrade WorldMirror, a feed-forward model for universal 3D prediction, by refining model architecture and learning strategy, enabling world reconstruction from multi-view images or videos. Also, we introduce WorldLens, a high-performance 3DGS rendering platform featuring a flexible engine-agnostic architecture, automatic IBL lighting, efficient collision detection, and training-rendering co-design, enabling interactive exploration of 3D worlds with character support. Extensive experiments demonstrate that HY-World 2.0 achieves state-of-the-art performance on several benchmarks among open-source approaches, delivering results comparable to the closed-source model Marble. We release all model weights, code, and technical details to facilitate reproducibility and support further research on 3D world models.

cs.CV↗

arXiv2026

Ultrawide-angle diffraction-limited 2D beam steering via hybrid integrated metasurface-photonic circuit

Two-dimensional (2D) wide field-of-view (FOV) beam steering is a key enabling capability for emerging free-space optical systems, including inter-satellite optical links, airborne LiDAR, point-to-point optical wireless communications, and collaborative robotic platforms. These applications require rapid acquisition and tracking across both azimuth and elevation; architectures that offer wide scanning in only one dimension while maintaining limited coverage in the orthogonal direction constrain link availability, coverage uniformity, and system agility. Here, we demonstrate a chip-scale platform for ultrawide-angle, diffraction-limited 2D beam steering based on hybrid integration of a silicon photonic integrated circuit (PIC) and an optical metasurface. A free-form micro-optical reflector efficiently transforms the guided waveguide mode into an expanded free-space beam that illuminates an analytically optimized ultrawide-FOV metasurface. The integrated system achieves a measured FOV exceeding 160° while maintaining diffraction-limited beam quality over a broad angular range at telecom wavelengths. This hybrid PIC-metasurface architecture provides a compact and scalable route to high-quality 2D beam steering and establishes a practical pathway toward integrated optical projectors for space-based optical communications and other applications requiring agile, wide-angle, high-fidelity beam control.

physics.optics↗