arXiv subjects

Yujia Zhang

Publications and source records attributed to Yujia Zhang.

At least 37 records · Page 2Linked to original sources

arXiv2026

A Spatial-Resolved Proton Energy Spectrometer Based on a Scintillation-Fiber Cube

Advanced particle acceleration methods have produced high-peak-current ion beams with broad energy spread and complex spatial distribution. There is an urgent need to develop online spatial-resolved energy spectrometers for high-energy pulsed ions. This paper introduces a novel spectrometer based on a scintillation-fiber cube for online diagnosis of proton beams with broadband energy spread and complex spatial distribution. We present its working principles, experimental setup, and comprehensive calibration using monoenergetic and spatially uniform proton beams generated by a synchrotron accelerator. Calibration results confirm an energy measurement range of 6-93 MeV, a relative energy uncertainty of 0.6% at 80 MeV, and a pixel size of 0.5 mm for beam profile reconstruction. By exploiting a custom-designed energy degrader, we generated a complex proton beam and measured it with the scintillation-fiber cube spectrometer (SFICS). The results demonstrate the spectrometer's potential for online measurement of the energy spectrum and spatial distribution of complex proton beams.

physics.acc-ph↗

arXiv2026

Power laws in the sea ice floe size distribution: a stochastic theory

Sea ice is a complex system, and observations have shown that ice segments (i.e., floes) have a wide range of sizes, with a floe size distribution that follows a power law. However, a theory for the power law and its exponent have remained elusive. Here, floe-resolving numerical simulations are investigated with a discrete element model, in order to gain further information by gathering statistics of fracture and welding events. Then, based on the insights from the floe-resolving simulations, a stochastic fragmentation-coagulation theory is proposed. Exact solutions are found with a power law. The power-law exponent can take a variety of values, and it depends on the fracture and welding rates. Such behavior is reminiscent of seasonal changes in the power-law exponent, which have been reported in past analyses of observational data.

physics.ao-ph↗

arXiv2026

Micro-Macro Retrieval: Reducing Long-Form Hallucination in Large Language Models

Large Language Models (LLMs) achieve impressive performance across many tasks but remain prone to hallucination, especially in long-form generation where redundant retrieved contexts and lengthy reasoning chains amplify factual errors. Recent studies highlight a critical phenomenon: the closer key information appears to the model outputs, the higher the factual accuracy. However, existing retrieval-augmented language models (RALMs) lack effective mechanisms to ensure this proximity - external evidence is injected into reasoning via multi-turn retrieval, but this cannot ensure key information stays close to the outputs. We propose Micro-Macro Retrieval (M2R), a novel retrieve-while-generate framework to fill this gap. At the macro level, M2R retrieves coarse-grained evidence from external sources; at the micro level, it extracts essential results from a key information repository built during reasoning and reuses them while generating answers. This design directly addresses the key-information-to-output proximity bottleneck, effectively reducing hallucination in long-form tasks. M2R is trained with a curriculum learning-based reinforcement learning strategy using customized rule-based rewards, enabling stable acquisition of retrieval and grounding skills. Extensive experiments across different benchmarks demonstrate the effectiveness of M2R, especially in lengthy-context settings.

cs.CL↗

arXiv2026

Concerto: Joint 2D-3D Self-Supervised Learning Emerges Spatial Representations

Humans learn abstract concepts through multisensory synergy, and once formed, such representations can often be recalled from a single modality. Inspired by this principle, we introduce Concerto, a minimalist simulation of human concept learning for spatial cognition, combining 3D intra-modal self-distillation with 2D-3D cross-modal joint embedding. Despite its simplicity, Concerto learns more coherent and informative spatial features, as demonstrated by zero-shot visualizations. It outperforms both standalone SOTA 2D and 3D self-supervised models by 14.2% and 4.8%, respectively, as well as their feature concatenation, in linear probing for 3D scene perception. With full fine-tuning, Concerto sets new SOTA results across multiple scene understanding benchmarks (e.g., 80.7% mIoU on ScanNet). We further present a variant of Concerto tailored for video-lifted point cloud spatial understanding, and a translator that linearly projects Concerto representations into CLIP's language space, enabling open-world perception. These results highlight that Concerto emerges spatial representations with superior fine-grained geometric and semantic consistency.

cs.CV↗

arXiv2026

MAPGD: Multi-Agent Prompt Gradient Descent for Collaborative Prompt Optimization

Prompt engineering is crucial for fully leveraging large language models (LLMs), yet most existing optimization methods follow a single trajectory, resulting in limited adaptability, gradient conflicts, and high computational overhead. We propose MAPGD (Multi-Agent Prompt Gradient Descent), a novel framework that reconceptualizes prompt optimization as a collaborative process among specialized agents. Each agent focuses on a distinct refinement dimension, such as instruction clarity, example selection, format structure, or stylistic adaptation, and their contributions are coordinated through semantic gradient embedding, conflict detection, and fusion. To further enhance robustness and stability, MAPGD introduces two new mechanisms: Hypersphere Constrained Gradient Clustering (HCGC), which enforces angular margin constraints for compact and well-separated clusters, and Channel Adaptive Agent Weighting (CAAW), which dynamically reweights agent contributions based on validation performance. Experiments on classification and reasoning benchmarks show that MAPGD consistently surpasses single-agent and random baselines in both accuracy and efficiency. Ablation studies confirm the effectiveness of gradient fusion, agent specialization, and conflict resolution. Together, these components establish MAPGD as a unified, gradient-based, and interpretable framework for robust prompt optimization with theoretical convergence guarantees.

cs.AI↗

arXiv2026

Bio-Inspired Photonic Spectral Encoders

Compact spectrometers promise to revolutionize sensing applications, offering a unique pathway to laboratory-grade analysis within a miniaturized footprint. Central to their performance is the encoding strategy to unknown spectra, which determines the efficiency, accuracy, and adaptability of spectral reconstruction. However, the absence of a unified spectral encoding framework has hindered the realization of optimal, high-performance compact spectrometers. We propose a transformative approach: an information-theoretic framework grounded in bio-inspired Bayesian expected information gain that defines the first generic light encoder for computational spectrometers. By optimizing three fundamental attributes at the lowest level of physical hierarchy, (1) orthogonality, (2) completeness, and (3) sparsity, we establish a design paradigm that transcends conventional encoding hardware limitations. We validate this paradigm with the first generic encoder capable of dynamically reconfiguring its response matrices. Experiments show superior reconstruction fidelity across diverse spectral regimes, enabling tunable spectral encoding tailored to varied input features. An ultra-high resolution of 6 pm and a broad measurable bandwidth of 30 nm are experimentally validated. By bridging the gap between theoretical encoding principles and reconfigurable hardware, our framework defines a coherent basis for future advances in compact spectrometry.

physics.optics↗

arXiv2026

Dressed-state relaxation in coupled qubits as a source of two-qubit gate errors

Understanding error mechanisms in two-qubit gate operations is essential for building high-fidelity quantum processors. While prior studies predominantly treat dephasing noise as either Markovian or predominantly low-frequency, realistic qubit environments exhibit structured, frequency-dependent spectra. Here we demonstrate that noise at frequencies matching the dressed-state energy splitting--set by the inter-qubit coupling strength g--induces a distinct relaxation channel that degrades gate performance. Through combined theoretical analysis and experimental verification on superconducting qubits with engineered noise spectra, we show that two-qubit gate errors scale predictably with the noise power spectral density at frequency 2g, extending the concept of $T_{1ρ}$ relaxation to interacting systems. This frequency-selective relaxation mechanism, universal across platforms, enriches our understanding of decoherence pathways during gate operations. The same mechanism sets coherence limits for dual-rail or singlet-triplet encodings.

quant-ph↗

arXiv2026

Suppressing spurious transitions using spectrally balanced pulse

Achieving precise control over quantum systems presents a significant challenge, especially in many-body setups, where residual couplings and unintended transitions undermine the accuracy of quantum operations. In superconducting qubits, parasitic interactions -- both between distant qubits and with spurious two-level systems -- can severely limit the performance of quantum gates. In this work, we introduce a pulse-shaping technique that uses spectrally balanced microwave pulses to suppress undesired transitions. Experimental results demonstrate an order-of-magnitude reduction in spurious excitations between weakly detuned qubits, as well as a substantial decrease in single-qubit gate errors caused by a strongly coupled two-level defect over a broad frequency range. Our method provides a simple yet powerful solution to mitigate adverse effects from parasitic couplings, enhancing the fidelity of quantum operations and expanding feasible frequency allocations for large-scale quantum devices.

quant-ph↗

arXiv2025

OmniPerson: Unified Identity-Preserving Pedestrian Generation

Person re-identification (ReID) suffers from a lack of large-scale high-quality training data due to challenges in data privacy and annotation costs. While previous approaches have explored pedestrian generation for data augmentation, they often fail to ensure identity consistency and suffer from insufficient controllability, thereby limiting their effectiveness in dataset augmentation. To address this, We introduce OmniPerson, the first unified identity-preserving pedestrian generation pipeline for visible/infrared image/video ReID tasks. Our contributions are threefold: 1) We proposed OmniPerson, a unified generation model, offering holistic and fine-grained control over all key pedestrian attributes. Supporting RGB/IR modality image/video generation with any number of reference images, two kinds of person poses, and text. Also including RGB-to-IR transfer and image super-resolution abilities.2) We designed Multi-Refer Fuser for robust identity preservation with any number of reference images as input, making OmniPerson could distill a unified identity from a set of multi-view reference images, ensuring our generated pedestrians achieve high-fidelity pedestrian generation.3) We introduce PersonSyn, the first large-scale dataset for multi-reference, controllable pedestrian generation, and present its automated curation pipeline which transforms public, ID-only ReID benchmarks into a richly annotated resource with the dense, multi-modal supervision required for this task. Experimental results demonstrate that OmniPerson achieves SoTA in pedestrian generation, excelling in both visual fidelity and identity consistency. Furthermore, augmenting existing datasets with our generated data consistently improves the performance of ReID models. We will open-source the full codebase, pretrained model, and the PersonSyn dataset.

cs.CV↗

arXiv2025

Ambiguity Awareness Optimization: Towards Semantic Disambiguation for Direct Preference Optimization

Direct Preference Optimization (DPO) is a widely used reinforcement learning from human feedback (RLHF) method across various domains. Recent research has increasingly focused on the role of token importance in improving DPO effectiveness. It is observed that identical or semantically similar content (defined as ambiguous content) frequently appears within the preference pairs. We hypothesize that the presence of ambiguous content during DPO training may introduce ambiguity, thereby limiting further improvements in alignment. Through mathematical analysis and proof-of-concept experiments, we reveal that ambiguous content may potentially introduce ambiguities, thereby degrading performance. To address this issue, we introduce Ambiguity Awareness Optimization (AAO), a simple yet effective approach that automatically re-weights ambiguous content to reduce ambiguities by calculating semantic similarity from preference pairs. Through extensive experiments, we demonstrate that AAO consistently and significantly surpasses state-of-the-art approaches in performance, without markedly increasing response length, across multiple model scales and widely adopted benchmark datasets, including AlpacaEval 2, MT-Bench, and Arena-Hard. Specifically, AAO outperforms DPO by up to 8.9 points on AlpacaEval 2 and achieves an improvement of by up to 15.0 points on Arena-Hard.

cs.CL↗

arXiv2025

A Preliminary Exploration of the Differences and Conjunction of Traditional PNT and Brain-inspired PNT

Developing universal Positioning, Navigation, and Timing (PNT) is our enduring goal. Today's complex environments demand PNT that is more resilient, energy-efficient and cognitively capable. This paper asks how we can endow unmanned systems with brain-inspired spatial cognition navigation while exploiting the high precision of machine PNT to advance universal PNT. We provide a new perspective and roadmap for shifting PNT from "tool-oriented" to "cognition-driven". Contributions: (1) multi-level dissection of differences among traditional PNT, biological brain PNT and brain-inspired PNT; (2) a four-layer (observation-capability-decision-hardware) fusion framework that unites numerical precision and brain-inspired intelligence; (3) forward-looking recommendations for future development of brain-inspired PNT.

cs.RO↗

arXiv2025

AIMMerging: Adaptive Iterative Model Merging Using Training Trajectories for Language Model Continual Learning

Continual learning (CL) is essential for deploying large language models (LLMs) in dynamic real-world environments without the need for costly retraining. Recent model merging-based methods have attracted significant attention, but they still struggle to effectively manage the trade-off between learning new knowledge and preventing forgetting, a challenge largely stemming from suboptimal number of merges and merging frequency. In this paper, we introduce Adaptive Iterative Model Merging (AimMerging), a novel CL framework that utilizes learning and forgetting signals from the training trajectory to dynamically monitor the model's training status. Guided by dynamic monitoring, the training trajectory-guided merge controller adaptively determines the timing and frequency of iterative fusion, while the rehearsal-based knowledge fusion module computes the merging weights and executes the fusion. Comprehensive experiments on three CL benchmarks with various model sizes (from 770M to 13B) demonstrate that AimMerging achieves significant performance improvements over existing state-of-the-art methods, with an average relative improvement of 80% and 59% on FWT and BWT, respectively. The source code is provided for reproducibility.

cs.CL↗

arXiv2025

ProPy: Building Interactive Prompt Pyramids upon CLIP for Partially Relevant Video Retrieval

Partially Relevant Video Retrieval (PRVR) is a practical yet challenging task that involves retrieving videos based on queries relevant to only specific segments. While existing works follow the paradigm of developing models to process unimodal features, powerful pretrained vision-language models like CLIP remain underexplored in this field. To bridge this gap, we propose ProPy, a model with systematic architectural adaption of CLIP specifically designed for PRVR. Drawing insights from the semantic relevance of multi-granularity events, ProPy introduces two key innovations: (1) A Prompt Pyramid structure that organizes event prompts to capture semantics at multiple granularity levels, and (2) An Ancestor-Descendant Interaction Mechanism built on the pyramid that enables dynamic semantic interaction among events. With these designs, ProPy achieves SOTA performance on three public datasets, outperforming previous models by significant margins. Code is available at https://github.com/BUAAPY/ProPy.

cs.CV↗

arXiv2025

A Low-Cost Portable Lidar-based Mobile Mapping System on an Android Smartphone

The rapid advancement of the metaverse, digital twins, and robotics underscores the demand for low-cost, portable mapping systems for reality capture. Current mobile solutions, such as the Leica BLK2Go and lidar-equipped smartphones, either come at a high cost or are limited in range and accuracy. Leveraging the proliferation and technological evolution of mobile devices alongside recent advancements in lidar technology, we introduce a novel, low-cost, portable mobile mapping system. Our system integrates a lidar unit, an Android smartphone, and an RTK-GNSS stick. Running on the Android platform, it features lidar-inertial odometry built with the NDK, and logs data from the lidar, wide-angle camera, IMU, and GNSS. With a total bill of materials (BOM) cost under 2,000 USD and a weight of about 1 kilogram, the system achieves a good balance between affordability and portability. We detail the system design, multisensor calibration, synchronization, and evaluate its performance for tracking and mapping. To further contribute to the community, the system's design and software are made open source at: https://github.com/OSUPCVLab/marslogger_android/releases/tag/v2.1

cs.RO↗

arXiv2025

Miniaturized Chaos-assisted Spectrometer

Computational spectrometers are at the forefront of spectroscopy, promising portable, on-chip, or in-situ spectrum analysis through the integration of advanced computational techniques into optical systems. However, existing computational spectrometer systems have yet to fully exploit optical properties due to imperfect spectral responses, resulting in increased system complexity and compromised performance in resolution, bandwidth, and footprint. In this study, we introduce optical chaos into spectrum manipulation via cavity deformation, leveraging high spatial and spectral complexities to address this challenge. By utilizing a single chaotic cavity, we achieve high diversity in spectra, facilitating channel decorrelation of 10 pm and ensuring optimal reconstruction over 100 nm within an ultra-compact footprint of 20*22 um2 as well as an ultra-low power consumption of 16.5 mW. Our approach not only enables state-of-the-art on-chip spectrometer performance in resolution-bandwidth-footprint metric, but also has the potential to revolutionize the entire computational spectrometer ecosystem.

physics.optics↗

arXiv2025

High computational density nanophotonic media for machine learning inference

Efficient machine learning inference is essential for the rapid adoption of artificial intelligence across various domains.On-chip optical computing has emerged as a transformative solution for accelerating machine learning tasks, owing to its ultra-low power consumption. However, enhancing the computational density of on-chip optical systems remains a significant challenge, primarily due to the difficulties in miniaturizing and integrating key optical interference components.In this work, we harness the potential of fabrication-constrained scattering optical computing within nanophotonic media to address these limitations.Central to our approach is the use of fabrication-aware inverse design techniques, which enable the realization of manufacturable on-chip scattering structures under practical constraints.This results in an ultra-compact optical neural computing architecture with an area of just 64 um2,representing a remarkable three orders of magnitude reduction in footprint compared to traditional optical neural networks. Our prototype, tested on the Iris flower dataset, achieved an experimental accuracy of 86.7%, closely matching the simulation benchmark.This breakthrough showcases a promising pathway toward ultra-dense, energy-efficient optical processors for scalable machine learning inference, significantly reducing both the hardware footprint, latency, and power consumption of next-generation AI applications.

physics.optics↗

arXiv2025

Correlation Improves Group Testing: Modeling Concentration-Dependent Test Errors

Population-wide screening is a powerful tool for controlling infectious diseases. Group testing enables such screening despite limited resources. Viral concentration of pooled samples are often positively correlated, either because prevalence and sample collection are influenced by location, or through intentional enhancement via pooling samples according to risk/household. Such correlation is known to improve efficiency under fixed test sensitivity. However, in reality, a test's sensitivity depends on the concentration of the analyte (e.g., viral RNA), as in the so-called dilution effect, where sensitivity decreases for larger pools. We show that concentration-dependent test error alters correlation's effect under the most widely-used group testing procedure, the two-stage Dorfman procedure. We prove that when test sensitivity increases with concentration, pooling correlated samples together (correlated pooling) achieves asymptotically higher sensitivity than independently pooling the samples (naive pooling). In contrast, in the concentration-independent case, correlation does not affect sensitivity. Moreover, with concentration-dependent errors, correlation can degrade test efficiency compared to naive pooling whereas under concentration-independent errors, correlation always improves efficiency. We propose an alternative measure of test resource usage, the number of positives found per test consumed, which we argue is better aligned with infection control, and show that correlated pooling outperforms naive pooling on this measure. In simulation, we show that the effect of correlation under realistic concentration-dependent test error meaningfully differs from correlation's effect assuming fixed sensitivity. Our findings underscore the importance for policy-makers of using models that incorporate naturally-occurring correlation and of considering ways of strengthening this correlation.

stat.AP↗

arXiv2025

Efficient Implementation of Arbitrary Two-Qubit Gates via Unified Control

The native gate set is fundamental to the performance of quantum devices, as it governs the accuracy of basic quantum operations and dictates the complexity of implementing quantum algorithms. Traditional approaches to extending gate sets often require accessing multiple transitions within an extended Hilbert space, leading to increased control complexity while offering only a limited set of gates. Here, we experimentally demonstrate a unified and highly versatile gate scheme capable of natively generating arbitrary two-qubit gates using only exchange interaction and qubit driving on a superconducting quantum processor, achieving maximum expressivity. Using a state-of-the-art transmon-coupler-transmon architecture, we achieve high fidelities averaging $99.37 \pm 0.07\%$ across a wide range of commonly used two-qubit unitaries. This outstanding performance, combined with reduced complexity, enables precise multipartite entangled state preparation, as demonstrated. To further enhance its applicability, we also show the high-fidelity realization of the unique B gate, which efficiently synthesizes the entire family of two-qubit gates. Our results highlight that fully exploiting the capabilities of a single interaction can yield a comprehensive and highly accurate gate set. With maximum expressivity, gate-time optimality, demonstrated high fidelity, and easy adaptability to other quantum platforms, our unified control scheme paves the way for optimal performance in quantum devices, offering exciting prospects for advancing quantum hardware and algorithm development.

quant-ph↗