arXiv subjects

Xi Feng

Publications and source records attributed to Xi Feng.

At least 19 recordsLinked to original sources

arXiv2026

Nanodiamond sensing in dynamic environments with fast-tracking through four-point positioning

Nitrogen-vacancy (NV) centers in nanodiamonds are excellent nanoscale sensors for measuring parameters such as temperature, magnetic field, and viscosity in complex fluidic environments, including living cells. However, the rapid motion of nanodiamonds in such dynamic systems imposes a significant challenge for continuous, real-time tracking and sensing measurements. Here, we present a fast single particle tracking (SPT) method featuring a tetrahedral detection geometry for time-efficient parallel fluorescence collection using four avalanche photodiodes (4-APDs), which eliminates the temporal latency of traditional sequential scanning. We demonstrate an improvement of about an order of magnitude in the temporal resolution and the upper limit of measurable diffusion coefficient compared to previously reported nanodiamond tracking methods based on single APD. The SPT is integrated with multi-parameter quantum sensing based on optically detected magnetic resonance (ODMR) of NV centers. The sensitivities of ODMR-based temperature and 3D rotation sensing are evaluated at different diffusion coefficients, which shows no significant degradation within our measurement range. We apply the system for thermorheology measurements in glycerol/water mixtures under thermal ramps. Additionally, we perform simultaneous translation and rotation tracking in live cells, revealing correlated translational and rotational dynamics. This approach advances multi-parameter nanoscale sensing for soft matter and biological applications, paving the way for real-time nanoscale sensing in highly dynamic fluidic environments.

physics.app-ph↗

arXiv2025

Electric-Magnetic-Switchable Free-Space Skyrmions in Toroidal Light Pulses via a Nonlinear Metasurface

Recent advances reveal that light propagation in free space supports many exotic topological textures, such as skyrmions. Their unique space-time topologies make them promising candidates as next-generation robust information carriers. Hence, the ability of switching different texture modes is highly demanded to serve as a manner of data transfer. However, previous studies focus on generation of one specific mode, lacking integrated devices with externally variable and stable mode generation capability. Here, we experimentally demonstrate the first realization of switchable skyrmions between electric and magnetic modes in toroidal light pulses using a nonlinear metasurface platform in terms of broadband terahertz generation driven by vectorial pulse. The spatial and temporal evolutions of them are also clearly observed. Our work establishes a new paradigm for manipulating and switching topologically structured light.

physics.optics↗

arXiv2025

RoboBrain 2.0 Technical Report

We introduce RoboBrain 2.0, our latest generation of embodied vision-language foundation models, designed to unify perception, reasoning, and planning for complex embodied tasks in physical environments. It comes in two variants: a lightweight 7B model and a full-scale 32B model, featuring a heterogeneous architecture with a vision encoder and a language model. Despite its compact size, RoboBrain 2.0 achieves strong performance across a wide spectrum of embodied reasoning tasks. On both spatial and temporal benchmarks, the 32B variant achieves leading results, surpassing prior open-source and proprietary models. In particular, it supports key real-world embodied AI capabilities, including spatial understanding (e.g., affordance prediction, spatial referring, trajectory forecasting) and temporal decision-making (e.g., closed-loop interaction, multi-agent long-horizon planning, and scene graph updating). This report details the model architecture, data construction, multi-stage training strategies, infrastructure and practical applications. We hope RoboBrain 2.0 advances embodied AI research and serves as a practical step toward building generalist embodied agents. The code, checkpoint and benchmark are available at https://superrobobrain.github.io.

cs.RO↗

arXiv2025

A Metric Framework for Triangle Inequalities via Barycentric Coordinates

This paper presents a unified metric-based framework for triangle geometric inequalities using barycentric coordinates. By interpreting classical inequalities as squared distances between points(a process termed metricization)we derive and refine numerous well-known inequalities. Furthermore, the squared-distance function ELD_I, measures distances from the incenter to points on the Euler line, also yielding generalized inequalities. The method offers geometric clarity and extends naturally to higher-dimensional and non-Euclidean settings.

math.MG↗

arXiv2025

Exact Closed-form Solutions for Lamb's Problem

In this article, we report on an exact closed-form solution for the displacement at the surface of an elastic half-space elicited by a buried point source that acts at some point underneath that surface. This is commonly referred to as the 3-D Lamb's problem, for which previous solutions were restricted to sources and receivers placed at the free surface. By means of the reciprocity theorem, our solution should also be valid as a means to obtain the displacements at interior points when the source is placed at the free surface. We manage to obtain explicit results by expressing the solution in terms of elementary algebraic expression as well as elliptic integrals. We anchor our developments on Poisson's ratio 0.25 starting from Johnson's (1974) integral solutions which must be computed numerically. In the end, our closed-form results agree perfectly with the numerical results of Johnson (1974), which strongly confirms the correctness of our explicit formulas. It is hoped that in due time, these formulas may constitute a valuable canonical solution that will serve as a yardstick against which other numerical solutions can be compared and measured.

physics.geo-ph↗

arXiv2025

Exact closed-form solutions for Lamb's problem (II): a moving point load

In this article, we report on an exact closed-form solution for the displacement in an elastic homogeneous half-space elicited by a downward vertical point source moving with constant velocity over the surface of the medium. The problem considered here is an extension to Lamb's problem. Starting with the integral solutions of Bakker \textit{et al.}, we followed the method developed in Feng and Zhang, which focuses on the displacement triggered by a fixed point source observed on the free surface, to obtain the final solution in terms of elementary algebraic functions as well as elliptic integrals of the first, second and third kind. Our closed-form results agree perfectly with the numerical results of Bakker \textit{et al.}, which confirms the correctness of our formulas. The solution obtained in this article may lay a solid foundation for further consideration of the response of an actual physical moving load, such as a high-speed rail train.

physics.geo-ph↗

arXiv2025

Exact closed-form solutions for Lamb's problem (III): The case for buried source and receiver

In this article, we derive the exact closed-form solution for the displacement in the interior of an elastic half-space due to a buried point force with Heaviside step function time history. It is referred to as the tensor Green's function for the elastic wave equation in a uniform half-space, also a natural generalization of the classical 3-D Lamb's problem, for which previous solutions have been restricted to the cases of either the source or the receiver or both are located on the free surface. Starting from the complex integral solutions of Johnson, we follow the similar procedures presented by Feng and Zhang to obtain the closed-form expressions in terms of elementary functions as well as elliptic integrals. Numerical results obtained from our closed-form expressions agree perfectly with those of Johnson, which validates our explicit formulae conclusively.

physics.geo-ph↗

arXiv2025

Metric Matrix in Barycentric Coordinates

In this paper, the concept of the metric matrix is introduced to establish a concise and unified formulation for the inner product in barycentric coordinates. Building on this framework, we explore the intrinsic algebraic identities of barycentric coordinates and their direct correspondence with geometric theorems. Through this approach, we derive a series of novel results and provide new proofs for several classical theorems in triangle geometry.

math.GM↗

arXiv2024

Photonic Terahertz Phased Array

Phased arrays are crucial in various technologies, such as radar and wireless communications, due to their ability to precisely control and steer electromagnetic waves. This precise control improves signal processing and enhances imaging performance. However, extending phased arrays to the terahertz (THz) frequency range has proven challenging, especially for high-frequency operation, broadband performance, two-dimensional (2D) phase control with large antenna arrays, and strong phase modulation. Here, we introduce a photonic platform to realize a THz phased array that bypasses the above challenges. Our method employs 2D phase coding with 2-bit across a broad THz frequency range from 0.8 to 1.4 THz. The core of our design is a pixelated nonlinear Pancharatnam-Berry metasurface driven by a spatially modulated femtosecond laser, allowing precise phase control of THz signals. We showcase the effectiveness of our method through four proof-of-concept applications: single beamforming, dual beamforming, imaging and vortex beam generation. The realized photonic platform provides a promising pathway for developing broadband phased arrays in the THz regime.

physics.optics↗

arXiv2024

AutoPatent: A Multi-Agent Framework for Automatic Patent Generation

As the capabilities of Large Language Models (LLMs) continue to advance, the field of patent processing has garnered increased attention within the natural language processing community. However, the majority of research has been concentrated on classification tasks, such as patent categorization and examination, or on short text generation tasks like patent summarization and patent quizzes. In this paper, we introduce a novel and practical task known as Draft2Patent, along with its corresponding D2P benchmark, which challenges LLMs to generate full-length patents averaging 17K tokens based on initial drafts. Patents present a significant challenge to LLMs due to their specialized nature, standardized terminology, and extensive length. We propose a multi-agent framework called AutoPatent which leverages the LLM-based planner agent, writer agents, and examiner agent with PGTree and RRAG to generate lengthy, intricate, and high-quality complete patent documents. The experimental results demonstrate that our AutoPatent framework significantly enhances the ability to generate comprehensive patents across various LLMs. Furthermore, we have discovered that patents generated solely with the AutoPatent framework based on the Qwen2.5-7B model outperform those produced by larger and more powerful LLMs, such as GPT-4o, Qwen2.5-72B, and LLAMA3.1-70B, in both objective metrics and human evaluations. We will make the data and code available upon acceptance at \url{https://github.com/QiYao-Wang/AutoPatent}.

cs.CL↗

arXiv2024

Can MLLMs Understand the Deep Implication Behind Chinese Images?

As the capabilities of Multimodal Large Language Models (MLLMs) continue to improve, the need for higher-order capability evaluation of MLLMs is increasing. However, there is a lack of work evaluating MLLM for higher-order perception and understanding of Chinese visual content. To fill the gap, we introduce the **C**hinese **I**mage **I**mplication understanding **Bench**mark, **CII-Bench**, which aims to assess the higher-order perception and understanding capabilities of MLLMs for Chinese images. CII-Bench stands out in several ways compared to existing benchmarks. Firstly, to ensure the authenticity of the Chinese context, images in CII-Bench are sourced from the Chinese Internet and manually reviewed, with corresponding answers also manually crafted. Additionally, CII-Bench incorporates images that represent Chinese traditional culture, such as famous Chinese traditional paintings, which can deeply reflect the model's understanding of Chinese traditional culture. Through extensive experiments on CII-Bench across multiple MLLMs, we have made significant findings. Initially, a substantial gap is observed between the performance of MLLMs and humans on CII-Bench. The highest accuracy of MLLMs attains 64.4%, where as human accuracy averages 78.2%, peaking at an impressive 81.0%. Subsequently, MLLMs perform worse on Chinese traditional culture images, suggesting limitations in their ability to understand high-level semantics and lack a deep knowledge base of Chinese traditional culture. Finally, it is observed that most models exhibit enhanced accuracy when image emotion hints are incorporated into the prompts. We believe that CII-Bench will enable MLLMs to gain a better understanding of Chinese semantics and Chinese-specific images, advancing the journey towards expert artificial general intelligence (AGI). Our project is publicly available at https://cii-bench.github.io/.

cs.CL↗

arXiv2024

II-Bench: An Image Implication Understanding Benchmark for Multimodal Large Language Models

The rapid advancements in the development of multimodal large language models (MLLMs) have consistently led to new breakthroughs on various benchmarks. In response, numerous challenging and comprehensive benchmarks have been proposed to more accurately assess the capabilities of MLLMs. However, there is a dearth of exploration of the higher-order perceptual capabilities of MLLMs. To fill this gap, we propose the Image Implication understanding Benchmark, II-Bench, which aims to evaluate the model's higher-order perception of images. Through extensive experiments on II-Bench across multiple MLLMs, we have made significant findings. Initially, a substantial gap is observed between the performance of MLLMs and humans on II-Bench. The pinnacle accuracy of MLLMs attains 74.8%, whereas human accuracy averages 90%, peaking at an impressive 98%. Subsequently, MLLMs perform worse on abstract and complex images, suggesting limitations in their ability to understand high-level semantics and capture image details. Finally, it is observed that most models exhibit enhanced accuracy when image sentiment polarity hints are incorporated into the prompts. This observation underscores a notable deficiency in their inherent understanding of image sentiment. We believe that II-Bench will inspire the community to develop the next generation of MLLMs, advancing the journey towards expert artificial general intelligence (AGI). II-Bench is publicly available at https://huggingface.co/datasets/m-a-p/II-Bench.

cs.CL↗

arXiv2024

CLHA: A Simple yet Effective Contrastive Learning Framework for Human Alignment

Reinforcement learning from human feedback (RLHF) is a crucial technique in aligning large language models (LLMs) with human preferences, ensuring these LLMs behave in beneficial and comprehensible ways to users. However, a longstanding challenge in human alignment techniques based on reinforcement learning lies in their inherent complexity and difficulty in training. To address this challenge, we present a simple yet effective Contrastive Learning Framework for Human Alignment (CLHA) to align LLMs with human preferences directly. CLHA employs a novel rescoring strategy to evaluate the noise within the data by considering its inherent quality and dynamically adjusting the training process. Simultaneously, CLHA utilizes pairwise contrastive loss and adaptive supervised fine-tuning loss to adaptively modify the likelihood of generating responses, ensuring enhanced alignment with human preferences. Using advanced methods, CLHA surpasses other algorithms, showcasing superior performance in terms of reward model scores, automatic evaluations, and human assessments on the widely used ``Helpful and Harmless'' dataset.

cs.AI↗

arXiv2021

Measurement of single-cell elasticity by nanodiamond-sensing of non-local deformation

Nano-indentation based on, e.g., atomic force microscopy (AFM), can measure single cell elasticity with high spatial resolution and sensitivity, but relating the data to cell mechanical properties depends on modeling that requires knowledge about the local contact between the indentation tip and the material, which is unclear in most cases. Here we use the orientation sensing by nitrogen-vacancy centers in nanodiamonds to chart the non-local deformation of fixed HeLa cells induced by AFM indentation, providing data for studying cell mechanics without requiring detailed knowledge about the local contact. The competition between the elasticity and capillarity on the cells is observed. We show that the apparent elastic moduli of the cells would have been overestimated if the capillarity is not considered (as in most previous studies using local depth-loading data). We also find reduction of both elastic moduli and surface tensions due to depolymerization of the actin cytoskeleton structure. This work demonstrates that, under shallow indentation, the nanodiamond sensing of non-local deformation with nanometer precision is particularly suitable for studying mechanics of cells.

physics.bio-ph↗

arXiv2021

Zero-field magnetometry using hyperfine-biased nitrogen-vacancy centers near diamond surfaces

Shallow nitrogen-vacancy (NV) centers in diamond are promising for nano-magnetometry for they can be placed proximate to targets. To study the intrinsic magnetic properties, zero-field magnetometry is desirable. However, for shallow NV centers under zero field, the strain near diamond surfaces would cause level anti-crossing between the spin states, leading to clock transitions whose frequencies are insensitive to magnetic signals. Furthermore, the charge noises from the surfaces would induce extra spin decoherence and hence reduce the magnetic sensitivity. Here we demonstrate that the relatively strong hyperfine coupling (130 MHz) from a first-shell 13C nuclear spin can provide an effective bias field to an NV center spin so that the clock-transition condition is broken and the charge noises are suppressed. The hyperfine bias enhances the dc magnetic sensitivity by a factor of 22 in our setup. With the charge noises suppressed by the strong hyperfine field, the ac magnetometry under zero field also reaches the limit set by decoherence due to the nuclear spin bath. In addition, the 130 MHz splitting of the NV center spin transitions allows relaxometry of magnetic noises simultaneously at two well-separated frequencies (~2.870 +/- 0.065 GHz), providing (low-resolution) spectral information of high-frequency noises under zero field. The hyperfine-bias enhanced zero-field magnetometry can be combined with dynamical decoupling to enhance single-molecule magnetic resonance spectroscopy and to improve the frequency resolution in nanoscale magnetic resonance imaging.

cond-mat.mes-hall↗

arXiv2019

Ultra-sensitive hybrid diamond nanothermometer

Nitrogen-vacancy (NV) centers in diamond are promising quantum sensors for their long spin coherence time under ambient conditions. However, their spin resonances are relatively insensitive to non-magnetic parameters such as temperature. A magnetic-nanoparticle-nanodiamond hybrid thermometer, where the temperature change is converted to the magnetic field variation near the Curie temperature, was demonstrated to have enhanced temperature sensitivity (11 mK Hz^{-1/2}) [Phys. Rev. X 8, 011042 (2018)], but the sensitivity was limited by the large spectral broadening of ensemble spins in nanodiamonds. To overcome this limitation, here we showed an improved design of a hybrid nanothermometer using a single NV center in a diamond nanopillar coupled with a single magnetic nanoparticle of copper-nickel alloy, and demonstrated a temperature sensitivity of 76 uK Hz^{-1/2}. This hybrid design enabled detection of 2 millikelvins temperature changes with temporal resolution of 5 milliseconds. The ultra-sensitive nanothermometer offers a new tool to investigate thermal processes in nanoscale systems.

quant-ph↗

arXiv2019

Nanometer-precision non-local deformation reconstruction using nanodiamond orientation sensing

Information of material deformation upon loading is critical to evaluate mechanical properties of materials in general, and key to understand fundamental mechano-stimuli induced response of live systems in particular. Conventionally, such information is obtained at macroscopic scale about averaged properties, which misses the important details of deformation at individual structures; or at nanoscale with localized properties accessed, which misses the information of deformation at locations away from the force-loading positions. Here we demonstrate that an integration of nanodiamond orientation sensing and AFM nanoindentation offers a complementary approach to the nonlocal reconstruction of material deformation. This approach features a 5 nm precision in the loading direction and a sub-hundred nanometer lateral resolution. Using this tool, we mapped out the deformation of a PDMS thin film in air and that of the gelatin microgel particles in water, with precision high enough to disclose the significance of the surface/interface effects in the material deformation. The non-local nanometer-precision sensing of deformation upon a local impact brings in new opportunities in studying mechanical response of complex material systems.

cond-mat.mes-hall↗

arXiv2018

Quantum coherence control at near 1000 K

Quantum coherence control usually requires extremely low temperature environments. Even for spins in diamond, a remarkable exception, the coherence signal is lost as temperature approaches 700 K. Here we demonstrate quantum coherence control of the electron spins of nitrogen-vacancy centers in nanodiamonds at temperatures near 1000 K. The scheme is based on initialization and readout of the spins at room temperature and control at high temperature, which is enabled by pulse laser heating and rapid diffusion cooling of nanodiamonds on amorphous carbon films. Using high-temperature spin control, we observe the magnetic phase transition of a single nickel nanoparticle at about 615 K. This work enables nano-thermometry and nano-magnetometry in the high-temperature regime.

quant-ph↗