arXiv subjects

Yubao Liu

Publications and source records attributed to Yubao Liu.

At least 19 recordsLinked to original sources

arXiv2026

Tomography of a Macroscopic Quantum State influenced by Classical Self-Gravity

Macroscopic optomechanical systems offer a promising testbed for distinguishing whether gravity acts as a quantum interaction or as a classical field. Schrodinger-Newton (SN) theory is the nonrelativistic limit of semi-classical gravity where quantum matter couples to classical gravity. Based on SN theory, this work investigates how classical self-gravity affects continuous quantum state tomography of a macroscopic mechanical oscillator monitored by variable-angle homodyne detection. In the Schrodinger-Newton (SN) theory, the measurement record arises from a different conditional test mass dynamics from that in quantum-gravity (QG)/standard quantum mechanics, consequently, applying the QG-optimised reconstruction map introduces an additional state-dependent contribution. We show that this contribution makes the reconstructed covariance depend on the chosen set of tomography angles and can drive the SN covariance--after QG filtering--outside the standard Gaussian-covariance domain set by the Heisenberg uncertainty principle. We quantify the resulting QG-SN distinguishability via the Hellinger distance and analyse its dependence on measurement strength and temperature. We then formulate the same issue in the broader setting of nonlinear quantum mechanics: when the system's conditional dynamics during the readout process depends on the state being inferred, the tomographic map acquires nonlinear, model-dependent corrections to the usual Radon or Gaussian reconstruction map.

quant-ph↗

arXiv2026

Efficient Traffic Forecasting on Large-Scale Road Network by Regularized Adaptive Graph Convolution

Traffic prediction is a critical task in spatial-temporal forecasting with broad applications in travel planning and urban management. To model the complex spatial-temporal dependencies in traffic data, Spatial-Temporal Graph Convolutional Networks (STGCNs) have been widely employed, achieving advanced performance. However, when applied to large-scale road networks, the quadratic computational complexity of traditional graph convolution operations severely limits their scalability. Several methods attempt to address this issue through approximation, compression, or spatial partitioning. Nevertheless, these methods often either fail to achieve sufficient computational efficiency or compromise prediction accuracy. To address these challenges, we propose a Regularized Adaptive Graph Convolution (RAGC) model. First, to ensure scalability on large road networks, we develop the Efficient Cosine Operator (ECO), which performs graph convolution based on the cosine similarity of node embeddings with linear time complexity. Second, we introduce a regularized adaptive graph convolution framework that combines Stochastic Shared Embedding (SSE) and adaptive graph convolution through a residual difference mechanism. This design enables the model to learn high-quality node embeddings, thereby improving prediction accuracy while maintaining computational efficiency. Extensive experiments on four large-scale real-world traffic datasets show that RAGC consistently outperforms state-of-the-art methods in terms of prediction accuracy and exhibits competitive computational efficiency. The code is available at: https://github.com/wkq-wukaiqi/RAGC.

cs.LG↗

arXiv2025

First result for testing semiclassical gravity effect with a torsion balance

The Schrödinger-Newton equation, a theoretical framework connecting quantum mechanics with classical gravity, predicts that gravity may induce measurable deviations in low-frequency mechanical systems-an intriguing hypothesis at the frontier of fundamental physics. In this study, we developed and operated an advanced optomechanical platform to investigate these effects. The system integrates an optical cavity with finesse over 350000 and a torsion pendulum with an ultra-low eigenfrequency of 0.6mHz, achieving a high mechanical Q-factor exceeding 50000. We collected data for 3 months and reached a sensitivity of 0.3urad/rtHz at the Schrödinger-Newton frequency of 2.5mHz where deviations from the standard quantum mechanics may occur. While no evidence supporting semiclassical gravity was found, we identify key challenges in such tests and propose new experimental approaches to advance this line of inquiry. This work demonstrates the potential of precision optomechanics to probe the interplay between quantum mechanics and gravity.

quant-ph↗

arXiv2025

Physically Consistent Global Atmospheric Data Assimilation with Machine Learning in Latent Space

Data assimilation (DA) integrates observations with model forecasts to produce optimized atmospheric states, whose physical consistency is critical for stable weather forecasting and reliable climate research. Traditional Bayesian DA methods enforce these nonlinear, flow-dependent physical constraints through empirical and tunable covariance structures, but with limited accuracy and robustness. Here, we introduce Latent Data Assimilation (LDA), a framework that performs Bayesian DA in a latent space learned from multivariate global atmospheric data via an autoencoder. We demonstrate that the autoencoder can largely capture nonlinear physical relationships, enabling LDA to produce balanced analyses without explicitly modeling physical constraints. Assimilation in latent space also improves both analysis quality and forecast skill compared to traditional model-space DA, under both idealized and real observational settings. Furthermore, LDA exhibits strong robustness across latent dimensions and remains effective even when the autoencoder is trained on inaccurate but physically realistic forecasts, highlighting its flexibility for real-world applications.

physics.ao-ph↗

arXiv2025

Testing the quantum nature of gravity through interferometry

We propose a Michelson-type interferometric protocol for testing the quantum nature of gravity through testing the phenomenology of semi-classical gravity theory, which predicts a state-dependent Schrodinger-Newton (SN) evolution of the test mass. The protocol's feature lies in utilizing the asymmetry of two interferometric arms induced by SN self-gravity to create cross-talk between the common and differential motion of the test masses. This cross-talk is imprinted as a clean binary signature in the correlation measurements of the interferometer's output light fields. Our results demonstrate that, when assisted by 10 dB squeezed input states, 3 hours of aggregated measurement data can provide sufficient signal-to-noise ratio to conclusively test the SN theory in 1 Kelvin environment. This shows the strong feasibility of using such interferometric protocols to test if gravity operates quantum-mechanically.

quant-ph↗

arXiv2025

Distinguishing Quantum and Classical Gravity via Non-Stationary Test Mass Dynamics

Classical gravity theory predicts a state-dependent gravitational potential for a quantum test mass, leading to nonlinear Schrodinger-Newton (SN) state evolution that contrasts with quantum gravity. Testing the effect of SN evolution can provide evidence for distinguishing quantum gravity and classical gravity, which is challenging to realize in the stationary optomechanical systems as analyzed in previous works [Phys. Rev. D 107, 024004 (2023), Phys. Rev. D 111, 062004 (2025)]. This work is devoted to analyzing the possibility of capturing the signature of SN theory during the non-stationary evolution of the test mass under the optomechanical measurement, where the second-order moments of a test mass can exhibit a distinctive oscillatory behavior. We show that this feature manifest in the non-stationary noise spectrum of outgoing light as additional peaks structures, although resolving these structures in practical experiments requires a larger number of repetitive trials with our sampling parameters, which is cost-prohibitive. To address this issue, we further employ statistical inference methods to extract more comprehensive information, thereby reducing the required number of experimental repetitions. Through Mock-Data simulations, we demonstrate that only 10 experimental trials of 40 seconds each are sufficient to reduce the false alarm rate for distinguishing between the two models to below one percent.

quant-ph↗

arXiv2025

GraphSparseNet: a Novel Method for Large Scale Traffic Flow Prediction

Traffic flow forecasting is a critical spatio-temporal data mining task with wide-ranging applications in intelligent route planning and dynamic traffic management. Recent advancements in deep learning, particularly through Graph Neural Networks (GNNs), have significantly enhanced the accuracy of these forecasts by capturing complex spatio-temporal dynamics. However, the scalability of GNNs remains a challenge due to their exponential growth in model complexity with increasing nodes in the graph. Existing methods to address this issue, including sparsification, decomposition, and kernel-based approaches, either do not fully resolve the complexity issue or risk compromising predictive accuracy. This paper introduces GraphSparseNet (GSNet), a novel framework designed to improve both the scalability and accuracy of GNN-based traffic forecasting models. GraphSparseNet is comprised of two core modules: the Feature Extractor and the Relational Compressor. These modules operate with linear time and space complexity, thereby reducing the overall computational complexity of the model to a linear scale. Our extensive experiments on multiple real-world datasets demonstrate that GraphSparseNet not only significantly reduces training time by 3.51x compared to state-of-the-art linear models but also maintains high predictive performance.

cs.LG↗

arXiv2025

The Role of Quantum Measurements when Testing the Quantum Nature of Gravity

In order to test the quantum nature of gravity, it is essential to explore the construction of classical gravity theories that are as consistent with experiments as possible. In particular, the classical gravity field must receive input regarding matter distribution. Previously, such input has been constructed by taking expectation values of the matter density operator or by using the outcomes of all measurements being performed on the quantum system. We propose a framework that unifies these models, and argue that the Causal Conditional Formulation of Schroedinger-Newton (CCSN) theory, which takes classical inputs only from experimental and environmental channels, is a minimum model within this framework. Since CCSN can be viewed as a quantum feedback control scheme, it can be made causal and free from pathologies that previously plagued SN theories. Since classical information from measurement results are used to generate classical gravity, CCSN can mimic quantum gravity better than one would naively expect for a classical theory. We predict experimental signatures of CCSN in two concrete scenarios: (i) a single test mass and (ii) two objects interacting via mutual gravity. In case (i), we show that the mass-concentration effect of self classical gravity still makes CCSN much easier to test than testing the mutual entanglement, yet the signatures are more subtle than previously thought for classical gravity theories. Using time-delayed and non-stationary measurements, which delay or suspend the flow of classical information into classical gravity, one can make CCSN more detectable. In case (ii), we show that mutual gravity generated by CCSN can lead to correlations that largely mimic signatures of quantum entanglement. Rigorous protocols that rule out LOCC channels, which are experimentally more challenging than simply testing entanglement, must be applied to completely rule out CCSN.

quant-ph↗

arXiv2024

Semiclassical gravity phenomenology under the causal-conditional quantum measurement prescription II: Heisenberg picture and apparent optical entanglement

The evolution of quantum states influenced by semiclassical gravity is distinct from that in quantum gravity theory due to the presence of a state-dependent gravitational potential. This state-dependent potential introduces nonlinearity into the state evolution, of which the theory is named Schroedinger-Newton (SN) theory. The formalism for understanding the continuous quantum measurement process on the quantum state in the context of semiclassical gravity theory has been previously discussed using the Schrödinger picture in Paper I [1]. In this work, an equivalent formalism using the Heisenberg picture is developed and applied to the analysis of two optomechanical experiment protocols that targeted testing the quantum nature of gravity. This Heisenberg picture formalism of the SN theory has the advantage of helping the investigation of the covariance matrices of the outgoing light fields in these protocols and further the entanglement features. We found that the classical gravity between the quantum trajectories of two mirrors under continuous quantum measurement in the SN theory can induce an apparent entanglement of the outgoing light field (though there is no quantum entanglement of the mirrors), which could serve as a false alarm for those experiments designed for probing the quantum gravity induced entanglement.

quant-ph↗

arXiv2022

Quantum scrambling in a toy model of photodetectors

Quantum measurement is a process that involves the interaction between a quantum system and a macroscopic measurement apparatus containing many degrees of freedom. The photodetector is such an apparatus with many electrons interacting with the incoming quantum photon. Therefore the incoming photon will spread and get scrambled in the photodetector, that is, the operator of the initial incoming local photons will grow and becomes highly non-local through the interaction process. Investigating this scrambling process in detail is useful for understanding the interaction between the quantum system and the measurement apparatus. In this paper, we study the quantum scrambling process in an effective toy model of photodetectors in three different physical scenarios, by numerically simulating the evolution of the out-of-time correlators (OTOC). In particular, the integrability of the effective model is explored through level spacing statistics, and the effect of the spatially/temporarily distributed disorders on the system evolution is carefully investigated using the OTOC. Looking into these detailed dynamical processes paves the way to the quantum simulation and manipulation of photondetector, which would provide insights into understanding the wave-function collapse processes.

quant-ph↗

arXiv2022

Semi-classical gravity phenomenology under the causal-conditional quantum measurement prescription

The semi-classical gravity sourced by the quantum expectation value of the matter's energy-momentum tensor will change the evolution of the quantum state of matter. This effect can be described by the Schroedinger-Newton (SN) equation, where the semi-classical gravity contributes a gravitational potential term depending on the matter quantum state. This state-dependent potential introduces the complexity of the quantum state evolution and measurement in SN theory, which is different for different quantum measurement prescriptions. Previous theoretical investigations on the SN-theory phenomenology in the optomechanical experimental platform were carried out under the so-called post/pre-selection prescription. This work will focus on the phenomenology of SN theory under the causal-conditional prescription, which fits the standard intuition on the continuous quantum measurement process. Under the causal-conditional prescription, the quantum state of the test mass mirrors is conditionally and continuously prepared by the projection of the outgoing light field in the optomechanical system. Therefore a gravitational potential depends on the quantum trajectory is created and further affects the system evolution. In this work, we will systematically study various experimentally measurable signatures of SN theory under the causal-conditional prescription in an optomechanical system, for both the self-gravity and the mutual gravity scenarios. Comparisons between the SN phenomenology under three different prescriptions will also be carefully made. Moreover, we find that quantum measurement can induce a classical correlation between two different optical fields via classical gravity, which is difficult to be distinguished from the quantum correlation of light fields mediated by quantum gravity.

quant-ph↗

arXiv2022

Scalable Simulation of Quantum Measurement Process with Quantum Computers

Recent development in quantum information sciences and technologies, especially building programmable quantum computers, provide us new opportunities to study fundamental aspects of quantum mechanics. We propose qubit models to emulate the quantum measurement process, in which the quantum information of a qubit is mapped to a collection of qubits acting as the measurement device. One model is motivated by single-photon detection and the other by spin measurement. Both models are scalable to generate Schrödinger cat-like state, and their corresponding quantum circuits are shown explicitly. Large-scale simulations could be realized in near-term quantum computers, while classical computers cannot perform the same task efficiently. Due to the scalability of the models, such simulations can help explore the quantum-to-classical boundary, if exists, in the quantum measurement problem. Besides, our protocol to generate cat states may have important applications in quantum computing and metrology.

quant-ph↗

arXiv2021

A Hybrid Rydberg Quantum Gate for Quantum Network

The high fidelity storage, distribution and processing of quantum information prefers qubits with different physical properties. Thus, hybrid quantum gates interfacing different types of qubits are essential for the realization of complex quantum network structures. A Rydberg-atom based physical quantum CZ gate is proposed to hybridly process the polarisation-encoded single-photon optical qubit and the "Schroedinger cat" microwave qubit. The degradation of the fidelity under the influence of various noise channels, such as microwave cavity loss, sponetanous emission of atom states, and non-adiabaticity effect, etc, has been analyised through detailed theoretical analysis by deriving input-output relation of qubit fields. The feasibility and the challenges of the protocol within current technology are also discussed by analysing the possible experimental parameter settings.

quant-ph↗

arXiv2020

Universal Model for Multi-Domain Medical Image Retrieval

Medical Image Retrieval (MIR) helps doctors quickly find similar patients' data, which can considerably aid the diagnosis process. MIR is becoming increasingly helpful due to the wide use of digital imaging modalities and the growth of the medical image repositories. However, the popularity of various digital imaging modalities in hospitals also poses several challenges to MIR. Usually, one image retrieval model is only trained to handle images from one modality or one source. When there are needs to retrieve medical images from several sources or domains, multiple retrieval models need to be maintained, which is cost ineffective. In this paper, we study an important but unexplored task: how to train one MIR model that is applicable to medical images from multiple domains? Simply fusing the training data from multiple domains cannot solve this problem because some domains become over-fit sooner when trained together using existing methods. Therefore, we propose to distill the knowledge in multiple specialist MIR models into a single multi-domain MIR model via universal embedding to solve this problem. Using skin disease, x-ray, and retina image datasets, we validate that our proposed universal model can effectively accomplish multi-domain MIR.

cs.CV↗

arXiv2020

Monitoring Depression Trend on Twitter during the COVID-19 Pandemic

The COVID-19 pandemic has severely affected people's daily lives and caused tremendous economic loss worldwide. However, its influence on people's mental health conditions has not received as much attention. To study this subject, we choose social media as our main data resource and create by far the largest English Twitter depression dataset containing 2,575 distinct identified depression users with their past tweets. To examine the effect of depression on people's Twitter language, we train three transformer-based depression classification models on the dataset, evaluate their performance with progressively increased training sizes, and compare the model's "tweet chunk"-level and user-level performances. Furthermore, inspired by psychological studies, we create a fusion classifier that combines deep learning model scores with psychological text features and users' demographic information and investigate these features' relations to depression signals. Finally, we demonstrate our model's capability of monitoring both group-level and population-level depression trends by presenting two of its applications during the COVID-19 pandemic. We hope this study can raise awareness among researchers and the general public of COVID-19's impact on people's mental health.

cs.SI↗

arXiv2019

Deeply Matting-based Dual Generative Adversarial Network for Image and Document Label Supervision

Although many methods have been proposed to deal with nature image super-resolution (SR) and get impressive performance, the text images SR is not good due to their ignorance of document images. In this paper, we propose a matting-based dual generative adversarial network (mdGAN) for document image SR. Firstly, the input image is decomposed into document text, foreground and background layers using deep image matting. Then two parallel branches are constructed to recover text boundary information and color information respectively. Furthermore, in order to improve the restoration accuracy of characters in output image, we use the input image's corresponding ground truth text label as extra supervise information to refine the two-branch networks during training. Experiments on real text images demonstrate that our method outperforms several state-of-the-art methods quantitatively and qualitatively.

cs.CV↗

arXiv2019

Neural Collective Entity Linking Based on Recurrent Random Walk Network Learning

Benefiting from the excellent ability of neural networks on learning semantic representations, existing studies for entity linking (EL) have resorted to neural networks to exploit both the local mention-to-entity compatibility and the global interdependence between different EL decisions for target entity disambiguation. However, most neural collective EL methods depend entirely upon neural networks to automatically model the semantic dependencies between different EL decisions, which lack of the guidance from external knowledge. In this paper, we propose a novel end-to-end neural network with recurrent random-walk layers for collective EL, which introduces external knowledge to model the semantic interdependence between different EL decisions. Specifically, we first establish a model based on local context features, and then stack random-walk layers to reinforce the evidence for related EL decisions into high-probability decisions, where the semantic interdependence between candidate entities is mainly induced from an external knowledge base. Finally, a semantic regularizer that preserves the collective EL decisions consistency is incorporated into the conventional objective function, so that the external knowledge base can be fully exploited in collective EL decisions. Experimental results and in-depth analysis on various datasets show that our model achieves better performance than other state-of-the-art models. Our code and data are released at \url{https://github.com/DeepLearnXMU/RRWEL}.

cs.CL↗

arXiv2016

Finding Multiple New Optimal Locations in a Road Network

We study the problem of optimal location querying for location based services in road networks, which aims to find locations for new servers or facilities. The existing optimal solutions on this problem consider only the cases with one new server. When two or more new servers are to be set up, the problem with minmax cost criteria, MinMax, becomes NP-hard. In this work we identify some useful properties about the potential locations for the new servers, from which we derive a novel algorithm for MinMax, and show that it is efficient when the number of new servers is small. When the number of new servers is large, we propose an efficient 3-approximate algorithm. We verify with experiments on real road networks that our solutions are effective and attains significantly better result quality compared to the existing greedy algorithms.

cs.DB↗