arXiv subjects

Yuhao Chen

Publications and source records attributed to Yuhao Chen.

At least 55 records · Page 3Linked to original sources

arXiv2025

Look as You Think: Unifying Reasoning and Visual Evidence Attribution for Verifiable Document RAG via Reinforcement Learning

Aiming to identify precise evidence sources from visual documents, visual evidence attribution for visual document retrieval-augmented generation (VD-RAG) ensures reliable and verifiable predictions from vision-language models (VLMs) in multimodal question answering. Most existing methods adopt end-to-end training to facilitate intuitive answer verification. However, they lack fine-grained supervision and progressive traceability throughout the reasoning process. In this paper, we introduce the Chain-of-Evidence (CoE) paradigm for VD-RAG. CoE unifies Chain-of-Thought (CoT) reasoning and visual evidence attribution by grounding reference elements in reasoning steps to specific regions with bounding boxes and page indexes. To enable VLMs to generate such evidence-grounded reasoning, we propose Look As You Think (LAT), a reinforcement learning framework that trains models to produce verifiable reasoning paths with consistent attribution. During training, LAT evaluates the attribution consistency of each evidence region and provides rewards only when the CoE trajectory yields correct answers, encouraging process-level self-verification. Experiments on vanilla Qwen2.5-VL-7B-Instruct with Paper- and Wiki-VISA benchmarks show that LAT consistently improves the vanilla model in both single- and multi-image settings, yielding average gains of 8.23% in soft exact match (EM) and 47.0% in IoU@0.5. Meanwhile, LAT not only outperforms the supervised fine-tuning baseline, which is trained to directly produce answers with attribution, but also exhibits stronger generalization across domains.

cs.AI↗

arXiv2025

The Solar Close Observations and Proximity Experiments (SCOPE) mission

The Solar Close Observations and Proximity Experiments (SCOPE) mission will send a spacecraft into the solar atmosphere at a low altitude of just 5 R_sun from the solar center. It aims to elucidate the mechanisms behind solar eruptions and coronal heating, and to directly measure the coronal magnetic field. The mission will perform in situ measurements of the current sheet between coronal mass ejections and their associated solar flares, and energetic particles produced by either reconnection or fast-mode shocks driven by coronal mass ejections. This will help to resolve the nature of reconnections in current sheets, and energetic particle acceleration regions. To investigate coronal heating, the mission will observe nano-flares on scales smaller than 70 km in the solar corona and regions smaller than 40 km in the photosphere, where magnetohydrodynamic waves originate. To study solar wind acceleration mechanisms, the mission will also track the process of ion charge-state freezing in the solar wind. A key achievement will be the observation of the coronal magnetic field at unprecedented proximity to the solar photosphere. The polar regions will also be observed at close range, and the inner edge of the solar system dust disk may be identified for the first time. This work presents the detailed background, science, and mission concept of SCOPE and discusses how we aim to address the questions mentioned above.

astro-ph.SR↗

arXiv2025

RGBSQGrasp: Inferring Local Superquadric Primitives from Single RGB Image for Graspability-Aware Bin Picking

Bin picking is a challenging robotic task due to occlusions and physical constraints that limit visual information for object recognition and grasping. Existing approaches often rely on known CAD models or prior object geometries, restricting generalization to novel or unknown objects. Other methods directly regress grasp poses from RGB-D data without object priors, but the inherent noise in depth sensing and the lack of object understanding make grasp synthesis and evaluation more difficult. Superquadrics (SQ) offer a compact, interpretable shape representation that captures the physical and graspability understanding of objects. However, recovering them from limited viewpoints is challenging, as existing methods rely on multiple perspectives for near-complete point cloud reconstruction, limiting their effectiveness in bin-picking. To address these challenges, we propose \textbf{RGBSQGrasp}, a grasping framework that leverages superquadric shape primitives and foundation metric depth estimation models to infer grasp poses from a monocular RGB camera -- eliminating the need for depth sensors. Our framework integrates a universal, cross-platform dataset generation pipeline, a foundation model-based object point cloud estimation module, a global-local superquadric fitting network, and an SQ-guided grasp pose sampling module. By integrating these components, RGBSQGrasp reliably infers grasp poses through geometric reasoning, enhancing grasp stability and adaptability to unseen objects. Real-world robotic experiments demonstrate a 92% grasp success rate, highlighting the effectiveness of RGBSQGrasp in packed bin-picking environments.

cs.RO↗

arXiv2025

SCALEX: Scalable Concept and Latent Exploration for Diffusion Models

Image generation models frequently encode social biases, including stereotypes tied to gender, race, and profession. Existing methods for analyzing these biases in diffusion models either focus narrowly on predefined categories or depend on manual interpretation of latent directions. These constraints limit scalability and hinder the discovery of subtle or unanticipated patterns. We introduce SCALEX, a framework for scalable and automated exploration of diffusion model latent spaces. SCALEX extracts semantically meaningful directions from H-space using only natural language prompts, enabling zero-shot interpretation without retraining or labelling. This allows systematic comparison across arbitrary concepts and large-scale discovery of internal model associations. We show that SCALEX detects gender bias in profession prompts, ranks semantic alignment across identity descriptors, and reveals clustered conceptual structure without supervision. By linking prompts to latent directions directly, SCALEX makes bias analysis in diffusion models more scalable, interpretable, and extensible than prior approaches.

cs.LG↗

arXiv2025

Assessing Power Flow Controllability via Variable Line Reactance

The rapid growth of large data center loads and inverter-based generation is increasing the stress on transmission networks, while expanding grid capacity at the required pace remains challenging. Power flow controllers (PFCs) that adjust effective line reactances to redistribute flows are often viewed as an interim solution to improve transmission network utilization. Traditional flexibility metrics and analysis approaches for PFCs focus on a limited number of operating points and contingencies. Towards gaining system-wide insights, this paper introduces a framework to quantify network flow controllability- the extent to which line flows can be reshaped through reactance adjustments. We derive analytical results demonstrating that installing PFCs on all lines enables complete controllability of feasible flow patterns. Building on these, we conduct empirical studies on the IEEE 39-bus system to examine how controllability varies with the number of PFCs and their reactance adjustment range. These analyses employ a mixed-integer linear program to optimize the siting and sizing of PFCs. Finally, we validate findings under AC power flow physics using an optimization routine that steers flows toward desired setpoints.

math.OC↗

arXiv2025

Xiangqi-R1: Enhancing Spatial Strategic Reasoning in LLMs for Chinese Chess via Reinforcement Learning

Game playing has long served as a fundamental benchmark for evaluating Artificial General Intelligence. While Large Language Models (LLMs) have demonstrated impressive capabilities in general reasoning, their effectiveness in spatial strategic reasoning, which is critical for complex and fully observable board games, remains insufficiently explored. In this work, we adopt Chinese Chess (Xiangqi) as a challenging and rich testbed due to its intricate rules and spatial complexity. To advance LLMs' strategic competence in such environments, we propose a training framework tailored to Xiangqi, built upon a large-scale dataset of five million board-move pairs enhanced with expert annotations and engine evaluations. Building on this foundation, we introduce Xiangqi-R1, a 7B-parameter model trained in multi-stage manner. Our Experimental results indicate that, despite their size and power, general-purpose LLMs struggle to achieve satisfactory performance in these tasks. Compared to general-purpose LLMs, Xiangqi-R1 greatly advances with an 18% rise in move legality and a 22% boost in analysis accuracy. Our results point to a promising path for creating general strategic intelligence in complex areas.

cs.AI↗

arXiv2025

A High-Precision Dynamical Model of Callisto: Incorporating Rotation Effects within Multi-Layer Internal Structure Models

China is planing to launch the Tianwen-4 mission around the year 2030, with its aim being the exploration of Jupiter and its moon, Callisto. Within the realm of deep space exploration, the accuracy of ephemerides is of great importance. Current ephemerides employ a simplified rotation model for Callisto, which this study addresses by proposing a novel dynamical model. This model enhancesthe existing orbital dynamics by integrating Callisto's rotational motions influenced by gravitational torques from the Sun, Jupiter, and other Galilean moons within an inertial frame, capturing the intricate coupling between Callisto's orbital and rotational dynamics. The study establishes a full dynamical model by deriving analytical expressions for this coupling and developing an adjustment model for data fitting using precise orbit determination methods. Furthermore, the influence of tidal effects on Callisto's motion is investigated, considering its multi-layered internal structure. Results demonstrate that the difference between the newly established full model and the model in current ephemerides is on the order of tens of meters. When calculating the impact of different internal structures of Callisto on its orbit, the influence of three-layered and two-layered structures is on the order of meters, suggesting that the development of a high-precision dynamical model requires additional constraints on the internal structure of Callisto. This research provides a novel alternative for a new generation of precise numerical ephemerides for Callisto. Additionally, these findings provide a testing platform for the data from the Tianwen-4 mission.

astro-ph.EP↗

arXiv2025

MosaicDoc: A Large-Scale Bilingual Benchmark for Visually Rich Document Understanding

Despite the rapid progress of Vision-Language Models (VLMs), their capabilities are inadequately assessed by existing benchmarks, which are predominantly English-centric, feature simplistic layouts, and support limited tasks. Consequently, they fail to evaluate model performance for Visually Rich Document Understanding (VRDU), a critical challenge involving complex layouts and dense text. To address this, we introduce DocWeaver, a novel multi-agent pipeline that leverages Large Language Models to automatically generate a new benchmark. The result is MosaicDoc, a large-scale, bilingual (Chinese and English) resource designed to push the boundaries of VRDU. Sourced from newspapers and magazines, MosaicDoc features diverse and complex layouts (including multi-column and non-Manhattan), rich stylistic variety from 196 publishers, and comprehensive multi-task annotations (OCR, VQA, reading order, and localization). With 72K images and over 600K QA pairs, MosaicDoc serves as a definitive benchmark for the field. Our extensive evaluation of state-of-the-art models on this benchmark reveals their current limitations in handling real-world document complexity and charts a clear path for future research.

cs.CV↗

arXiv2025

FLAD: Federated Learning for LLM-based Autonomous Driving in Vehicle-Edge-Cloud Networks

Large Language Models (LLMs) have impressive data fusion and reasoning capabilities for autonomous driving (AD). However, training LLMs for AD faces significant challenges including high computation transmission costs, and privacy concerns associated with sensitive driving data. Federated Learning (FL) is promising for enabling autonomous vehicles (AVs) to collaboratively train models without sharing raw data. We present Federated LLM-based Autonomous Driving (FLAD), an FL framework that leverages distributed multimodal sensory data across AVs in heterogeneous environment. FLAD has three key innovations: (1) a cloud-edge-vehicle collaborative architecture that reduces communication delay and preserving data privacy; (2) an intelligent parallelized collaborative training with a communication scheduling mechanism that optimizes training efficiency, leveraging end-devices otherwise having insufficient resources for model training; and (3) a knowledge distillation method that personalizes LLM according to heterogeneous edge data. In addition, we prototype FLAD in a testbed with NVIDIA Jetsons, overcoming practical implementation challenges including CPU/GPU memory sharing in resource-constrained devices, dynamic model partitions, and fault-tolerant training.Extensive experimental evaluation demonstrates that FLAD achieves superior end-to-end AD performance while efficiently utilizing distributed vehicular resources, opening up new possibilities for future collaborative AD model training and knowledge sharing.

cs.LG↗

arXiv2025

DreamPose3D: Hallucinative Diffusion with Prompt Learning for 3D Human Pose Estimation

Accurate 3D human pose estimation remains a critical yet unresolved challenge, requiring both temporal coherence across frames and fine-grained modeling of joint relationships. However, most existing methods rely solely on geometric cues and predict each 3D pose independently, which limits their ability to resolve ambiguous motions and generalize to real-world scenarios. Inspired by how humans understand and anticipate motion, we introduce DreamPose3D, a diffusion-based framework that combines action-aware reasoning with temporal imagination for 3D pose estimation. DreamPose3D dynamically conditions the denoising process using task-relevant action prompts extracted from 2D pose sequences, capturing high-level intent. To model the structural relationships between joints effectively, we introduce a representation encoder that incorporates kinematic joint affinity into the attention mechanism. Finally, a hallucinative pose decoder predicts temporally coherent 3D pose sequences during training, simulating how humans mentally reconstruct motion trajectories to resolve ambiguity in perception. Extensive experiments on benchmarked Human3.6M and MPI-3DHP datasets demonstrate state-of-the-art performance across all metrics. To further validate DreamPose3D's robustness, we tested it on a broadcast baseball dataset, where it demonstrated strong performance despite ambiguous and noisy 2D inputs, effectively handling temporal consistency and intent-driven motion variations.

cs.CV↗

arXiv2025

MGSO: Monocular Real-time Photometric SLAM with Efficient 3D Gaussian Splatting

Real-time SLAM with dense 3D mapping is computationally challenging, especially on resource-limited devices. The recent development of 3D Gaussian Splatting (3DGS) offers a promising approach for real-time dense 3D reconstruction. However, existing 3DGS-based SLAM systems struggle to balance hardware simplicity, speed, and map quality. Most systems excel in one or two of the aforementioned aspects but rarely achieve all. A key issue is the difficulty of initializing 3D Gaussians while concurrently conducting SLAM. To address these challenges, we present Monocular GSO (MGSO), a novel real-time SLAM system that integrates photometric SLAM with 3DGS. Photometric SLAM provides dense structured point clouds for 3DGS initialization, accelerating optimization and producing more efficient maps with fewer Gaussians. As a result, experiments show that our system generates reconstructions with a balance of quality, memory efficiency, and speed that outperforms the state-of-the-art. Furthermore, our system achieves all results using RGB inputs. We evaluate the Replica, TUM-RGBD, and EuRoC datasets against current live dense reconstruction systems. Not only do we surpass contemporary systems, but experiments also show that we maintain our performance on laptop hardware, making it a practical solution for robotics, A/R, and other real-time applications.

cs.RO↗

arXiv2025

Coronal Mass Ejections Deflected by Newly Emerging Flux: A Combined Analytic and Numerical Study

Newly emerging flux (NEF) has been widely studied as a trigger of solar filament eruptions, but its influence on the subsequent dynamics remains poorly explored. Because NEF typically emerges adjacent to filaments, it imposes magnetic asymmetry that can drive non-radial eruptions and complicate space-weather forecasting. We bridge analytic catastrophe theory with 2D resistive MHD simulations: analytic solutions provide magnetic configurations containing a flux rope at the loss-of-equilibrium point, which are then used as initial conditions for simulations to examine the following dynamics. We find that NEF governs the kinematics of filament eruptions in two ways. First, by reshaping coronal stability, NEF can create or eliminate a higher equilibrium in corona, thereby producing failed eruptions or CMEs. In the transitional situation where a metastable equilibrium appears, the rising filament decelerates and stalls before re-accelerating into a CME, consistent with observed two-step eruptions. Second, by breaking symmetry, NEF deflects eruptions away from the radial direction: depending on its polarity, it acts as a repulsor or an attractor on eruptive filaments, and the deflection magnitude increases with the degree of asymmetry. Our theory yields two characteristic angles that predict the deflection directions of CMEs and failed eruptions, and simulations closely aligns with these predictors. These results highlight the NEF not only as a trigger but also as a key factor that governs both the acceleration and deflection of eruptions during their propagation in the low corona.

astro-ph.SR↗

arXiv2025

QFP Waves Driven by the Tuning-Fork Effect during Magnetic Reconnecion

Through three-dimensional MHD simulations, we have uncovered a kind of fast coronal wave originating from both ends of a current sheet (CS) during a solar eruption. These waves are observed to appear near the top and bottom ends of the reconnection-related CS. The simulations demonstrate the presence of termination shock regions above the two ends of the CS. As the reconnection outflows escape from the vertical CS and encounter these termination shocks, they undergo partial reflection, redirecting towards the CS terminal fork walls. The identified waves propagate rapidly at a speed of approximately 1400 km/s with a period of just 2 s. Concurrently, the time-evolution of intensity within a small region of the CS terminal fork structures, exhibits a similar oscillation period of 2 s. All these evidence supports the notion that these QFP (Quasi-periodic Fast-Propagating) waves were excited by tuning fork effects within the CS system. Essentially, the rapid reconnection outflows are reflected by the terminal shocks, striking the fork walls at the CS ends. Moreover, parts of the oscillations along the tuning fork handle are transformed into thermal energy, accumulating in the CS center and elevating the temperature. This is the first time to report such QFP waves resulting from tuning fork effects within the CS during a solar eruption. These waves are anticipated to manifest closely following the propagation of CMEs and adjacent to the related post-flare loops in observations, with partial confirmation in current observations.

astro-ph.SR↗

arXiv2025

Say It Another Way: Auditing LLMs with a User-Grounded Automated Paraphrasing Framework

Large language models (LLMs) are highly sensitive to subtle changes in prompt phrasing, posing challenges for reliable auditing. Prior methods often apply unconstrained prompt paraphrasing, which risk missing linguistic and demographic factors that shape authentic user interactions. We introduce AUGMENT (Automated User-Grounded Modeling and Evaluation of Natural Language Transformations), a framework for generating controlled paraphrases, grounded in user behaviors. AUGMENT leverages linguistically informed rules and enforces quality through checks on instruction adherence, semantic similarity, and realism, ensuring paraphrases are both reliable and meaningful for auditing. Through case studies on the BBQ and MMLU datasets, we show that controlled paraphrases uncover systematic weaknesses that remain obscured under unconstrained variation. These results highlight the value of the AUGMENT framework for reliable auditing.

cs.CL↗

arXiv2025

Specific multi-emitter identification via multi-label learning

Specific emitter identification leverages hardware-induced impairments to uniquely determine a specific transmitter. However, existing approaches fail to address scenarios where signals from multiple emitters overlap. In this paper, we propose a specific multi-emitter identification (SMEI) method via multi-label learning to determine multiple transmitters. Specifically, the multi-emitter fingerprint extractor is designed to mitigate the mutual interference among overlapping signals. Then, the multi-emitter decision maker is proposed to assign the all emitter identification using the previous extracted fingerprint. Experimental results demonstrate that, compared with baseline approach, the proposed SMEI scheme achieves comparable identification accuracy under various overlapping conditions, while operating at significantly lower complexity. The significance of this paper is to identify multiple emitters from overlapped signal with a low complexity.

eess.SP↗

arXiv2025

LEDiT: Your Length-Extrapolatable Diffusion Transformer without Positional Encoding

Diffusion transformers (DiTs) struggle to generate images at resolutions higher than their training resolutions. The primary obstacle is that the explicit positional encodings(PE), such as RoPE, need extrapolating to unseen positions which degrades performance when the inference resolution differs from training. In this paper, We propose a Length-Extrapolatable Diffusion Transformer~(LEDiT) to overcome this limitation. LEDiT needs no explicit PEs, thereby avoiding PE extrapolation. The key innovation of LEDiT lies in the use of causal attention. We demonstrate that causal attention can implicitly encode global positional information and show that such information facilitates extrapolation. We further introduce a locality enhancement module, which captures fine-grained local information to complement the global coarse-grained position information encoded by causal attention. Experimental results on both conditional and text-to-image generation tasks demonstrate that LEDiT supports up to 4x resolution scaling (e.g., from 256x256 to 512x512), achieving better image quality compared to the state-of-the-art length extrapolation methods. We believe that LEDiT marks a departure from the standard RoPE-based methods and offers a promising insight into length extrapolation. Project page: https://shenzhang2145.github.io/ledit/

cs.CV↗

arXiv2025

Optimally Linearizing Power Flow Equations for Improved Power System Dispatch

Managing power grids with the increasing presence of variable renewable energy-based (distributed) generation involves solving high-dimensional optimization tasks at short intervals. Linearizing the AC power flow (PF) constraints is a standard practice to ease the computational burden at the cost of hopefully acceptable inaccuracies. However, the design of these PF linearizations has traditionally been agnostic of the use case. Towards bridging the linearization-application gap, we first model the complete operational sequence needed to implement optimal power flow (OPF) decisions on power systems and characterize the effect of PF linearization on the resulting steady-state system operation. We then propose a novel formulation for obtaining optimal PF constraint linearizations to harness desirable system-operation attributes such as low generation cost and engineering-limit violations. To pursue the optimal PF linearization, we develop a gradient-based approach backed by sensitivity analysis of optimization routines and AC PF equations. Numerical tests on the IEEE 39-bus system demonstrate the capabilities of our approach in traversing the cost-optimality vs operational feasibility trade-off inherent to OPF approximations.

math.OC↗

arXiv2025

GS: Generative Segmentation via Label Diffusion

Language-driven image segmentation is a fundamental task in vision-language understanding, requiring models to segment regions of an image corresponding to natural language expressions. Traditional methods approach this as a discriminative problem, assigning each pixel to foreground or background based on semantic alignment. Recently, diffusion models have been introduced to this domain, but existing approaches remain image-centric: they either (i) use image diffusion models as visual feature extractors, (ii) synthesize segmentation data via image generation to train discriminative models, or (iii) perform diffusion inversion to extract attention cues from pre-trained image diffusion models-thereby treating segmentation as an auxiliary process. In this paper, we propose GS (Generative Segmentation), a novel framework that formulates segmentation itself as a generative task via label diffusion. Instead of generating images conditioned on label maps and text, GS reverses the generative process: it directly generates segmentation masks from noise, conditioned on both the input image and the accompanying language description. This paradigm makes label generation the primary modeling target, enabling end-to-end training with explicit control over spatial and semantic fidelity. To demonstrate the effectiveness of our approach, we evaluate GS on Panoptic Narrative Grounding (PNG), a representative and challenging benchmark for multimodal segmentation that requires panoptic-level reasoning guided by narrative captions. Experimental results show that GS significantly outperforms existing discriminative and diffusion-based methods, setting a new state-of-the-art for language-driven segmentation.

cs.CV↗