arXiv subjects

Siyuan Chen

Publications and source records attributed to Siyuan Chen.

At least 109 records · Page 6Linked to original sources

arXiv2024

Plasticine3D: 3D Non-Rigid Editing with Text Guidance by Multi-View Embedding Optimization

With the help of Score Distillation Sampling (SDS) and the rapid development of neural 3D representations, some methods have been proposed to perform 3D editing such as adding additional geometries, or overwriting textures. However, generalized 3D non-rigid editing task, which requires changing both the structure (posture or composition) and appearance (texture) of the original object, remains to be challenging in 3D editing field. In this paper, we propose Plasticine3D, a novel text-guided fine-grained controlled 3D editing pipeline that can perform 3D non-rigid editing with large structure deformations. Our work divides the editing process into a geometry editing stage and a texture editing stage to achieve separate control of structure and appearance. In order to maintain the details of the original object from different viewpoints, we propose a Multi-View-Embedding (MVE) Optimization strategy to ensure that the guidance model learns the features of the original object from various viewpoints. For the purpose of fine-grained control, we propose Embedding-Fusion (EF) to blend the original characteristics with the editing objectives in the embedding space, and control the extent of editing by adjusting the fusion rate. Furthermore, in order to address the issue of gradual loss of details during the generation process under high editing intensity, as well as the problem of insignificant editing effects in some scenarios, we propose Score Projection Sampling (SPS) as a replacement of score distillation sampling, which introduces additional optimization phases for editing target enhancement and original detail maintenance, leading to better editing quality. Extensive experiments demonstrate the effectiveness of our method on 3D non-rigid editing tasks

cs.CV↗

arXiv2024

Mobile-Env: Building Qualified Evaluation Benchmarks for LLM-GUI Interaction

The Graphical User Interface (GUI) is pivotal for human interaction with the digital world, enabling efficient device control and the completion of complex tasks. Recent progress in Large Language Models (LLMs) and Vision Language Models (VLMs) offers the chance to create advanced GUI agents. To ensure their effectiveness, there's a pressing need for qualified benchmarks that provide trustworthy and reproducible evaluations -- a challenge current benchmarks often fail to address. To tackle this issue, we introduce Mobile-Env, a comprehensive toolkit tailored for creating GUI benchmarks in the Android mobile environment. Mobile-Env offers an isolated and controllable setting for reliable evaluations, and accommodates intermediate instructions and rewards to reflect real-world usage more naturally. Utilizing Mobile-Env, we collect an open-world task set across various real-world apps and a fixed world set, WikiHow, which captures a significant amount of dynamic online contents for fully controllable and reproducible evaluation. We conduct comprehensive evaluations of LLM agents using these benchmarks. Our findings reveal that even advanced models (e.g., GPT-4V and LLaMA-3) struggle with tasks that are relatively simple for humans. This highlights a crucial gap in current models and underscores the importance of developing more capable foundation models and more effective GUI agent frameworks.

cs.AI↗

arXiv2024

PsyEval: A Suite of Mental Health Related Tasks for Evaluating Large Language Models

Evaluating Large Language Models (LLMs) in the mental health domain poses distinct challenged from other domains, given the subtle and highly subjective nature of symptoms that exhibit significant variability among individuals. This paper presents PsyEval, the first comprehensive suite of mental health-related tasks for evaluating LLMs. PsyEval encompasses five sub-tasks that evaluate three critical dimensions of mental health. This comprehensive framework is designed to thoroughly assess the unique challenges and intricacies of mental health-related tasks, making PsyEval a highly specialized and valuable tool for evaluating LLM performance in this domain. We evaluate twelve advanced LLMs using PsyEval. Experiment results not only demonstrate significant room for improvement in current LLMs concerning mental health but also unveil potential directions for future model optimization.

cs.CL↗

arXiv2024

Status Report on Global Pulsar-Timing-Array Efforts to Detect Gravitational Waves

The stability of the spin of pulsars and the precision with which these spins can be determined, allows many unique tests of interest to physics and astrophysics. Perhaps the most challenging and revolutionary of these, is the detection of nanohertz gravitational waves. An increasing number of efforts to detect and study long-period gravitational waves by timing an array of pulsars have been ongoing for several decades and the field is moving ever closer to actual gravitational-wave science. In this review article, we summarise the state of this field by presenting the current sensitivity to gravitational waves and by reviewing recent progress along the multiple lines of research that are part of the continuous push towards greater sensitivity. We also briefly review some of the most recent efforts at astrophysical interpretation of the most recent GW estimates derived from pulsar timing.

astro-ph.HE↗

arXiv2024

Mixed-Precision Computing in the GRIST Dynamical Core for Weather and Climate Modelling

Atmosphere modelling applications become increasingly memory-bound due to the inconsistent development rates between processor speeds and memory bandwidth. In this study, we mitigate memory bottlenecks and reduce the computational load of the GRIST dynamical core by adopting the mixed-precision computing strategy. Guided by a limited-degree of iterative development principle, we identify the equation terms that are precision insensitive and modify them from double- to single-precision. The results show that most precision-sensitive terms are predominantly linked to pressure-gradient and gravity terms, while most precision-insensitive terms are advective terms. The computational cost is reduced without compromising the solver accuracy. The runtime of the model's hydrostatic solver, non-hydrostatic solver, and tracer transport solver is reduced by 24%, 27%, and 44%, respectively. A series of idealized tests, real-world weather and climate modelling tests, has been performed to assess the optimized model performance qualitatively and quantitatively. In particular, in the high-resolution weather forecast simulation, the model sensitivity to the precision level is mainly dominated by the small-scale features. While in long-term climate simulation, the precision-induced sensitivity can form at the large scale.

physics.ao-ph↗

arXiv2024

A Lane Usage Strategy for General Traffic Access on Bus Lanes under Mixed Traffic Environment

The strategy of permitting general traffic to use the bus lane for improved utilization while ensuring bus priority has gained increasingly attention, particularly with the support of vehicle-to-everything technology. In this study, we propose a novel lane usage strategy called Dynamic Spatial-Temporal Priority (DSTP) to ensure bus priority and optimize bus lane usage in a mixed traffic environment. DSTP leverages dynamic methods to identify available spatial-temporal resources in the lane, utilizing signal timing, road information, and vehicle data. A Right-of-Way assignment optimization model is then developed based on these resources to determine which vehicles can enter the bus lane. The model is dynamically enacted using a rolling horizon scheme to accommodate time-varying traffic conditions. Numerical studies have validated the advantages of DSTP, showing maintained bus priority, improved traffic efficiency, reduced fuel consumption, and lower CO2 emissions, especially during periods of high traffic demand and concentrated bus arrivals.

math.OC↗

arXiv2024

Mass-redshift dependency of Supermassive Black Hole Binaries for the Gravitational Wave Background

Studying how the black hole (BH) - (galaxy) bulge mass relation evolves with redshift provides valuable insights into the co-evolution of supermassive black holes and their host galaxies. However, obtaining accurate measurement of BH masses is challenging due to the bias towards the most massive and luminous galaxies. Instead we focus on the BH and bulge masses as they vary with redshift using the EAGLE, Illustris, TNG100, TNG300, Horizon-AGN and SIMBA large-scale cosmological simulations. We use an analytical astrophysical model with galaxy stellar mass function, pair fraction, merger timescale and BH-bulge mass relation extended to include redshift evolution. The model can predict the intensity of the gravitational wave background (GWB) produced by a population of supermassive black hole binary (SMBHB) as a function of the frequency. This allows us to compare the predictions of this model with the constraints of Pulsar Timing Array observations. Here, we employ Bayesian analysis for the parameter inference. We find that all six simulations are consistent $\leq 3.5σ$ with a range of simulated GWB spectra. By fixing the BH-bulge mass parameters to the simulations we analyze the changes in the constraints on the other astrophysical parameters. Furthermore, we also examine the variation in SMBHB merger rate with mass and redshift between these large-scale simulations.

astro-ph.CO↗

arXiv2024

EyeEcho: Continuous and Low-power Facial Expression Tracking on Glasses

In this paper, we introduce EyeEcho, a minimally-obtrusive acoustic sensing system designed to enable glasses to continuously monitor facial expressions. It utilizes two pairs of speakers and microphones mounted on glasses, to emit encoded inaudible acoustic signals directed towards the face, capturing subtle skin deformations associated with facial expressions. The reflected signals are processed through a customized machine-learning pipeline to estimate full facial movements. EyeEcho samples at 83.3 Hz with a relatively low power consumption of 167 mW. Our user study involving 12 participants demonstrates that, with just four minutes of training data, EyeEcho achieves highly accurate tracking performance across different real-world scenarios, including sitting, walking, and after remounting the devices. Additionally, a semi-in-the-wild study involving 10 participants further validates EyeEcho's performance in naturalistic scenarios while participants engage in various daily activities. Finally, we showcase EyeEcho's potential to be deployed on a commercial-off-the-shelf (COTS) smartphone, offering real-time facial expression tracking.

cs.HC↗

arXiv2024

GazeTrak: Exploring Acoustic-based Eye Tracking on a Glass Frame

In this paper, we present GazeTrak, the first acoustic-based eye tracking system on glasses. Our system only needs one speaker and four microphones attached to each side of the glasses. These acoustic sensors capture the formations of the eyeballs and the surrounding areas by emitting encoded inaudible sound towards eyeballs and receiving the reflected signals. These reflected signals are further processed to calculate the echo profiles, which are fed to a customized deep learning pipeline to continuously infer the gaze position. In a user study with 20 participants, GazeTrak achieves an accuracy of 3.6° within the same remounting session and 4.9° across different sessions with a refreshing rate of 83.3 Hz and a power signature of 287.9 mW. Furthermore, we report the performance of our gaze tracking system fully implemented on an MCU with a low-power CNN accelerator (MAX78002). In this configuration, the system runs at up to 83.3 Hz and has a total power signature of 95.4 mW with a 30 Hz FPS.

cs.HC↗

arXiv2024

A Gaussian-processes approach to fitting for time-variable spherical solar wind in pulsar timing data

Propagation effects are one of the main sources of noise in high-precision pulsar timing. For pulsars below an ecliptic latitude of $5^\circ$, the ionised plasma in the solar wind can introduce dispersive delays of order 100 microseconds around solar conjunction at an observing frequency of 300 MHz. A common approach to mitigate this assumes a spherical solar wind with a time-constant amplitude. However, this has been shown to be insufficient to describe the solar wind. We present a linear, Gaussian-process piecewise Bayesian approach to fit a spherical solar wind of time-variable amplitude, which has been implemented in the pulsar software run_enterprise. Through simulations, we find that the current EPTA+InPTA data combination is not sensitive to such variations; however, solar wind variations will become important in the near future with the addition of new InPTA data and data collected with the low-frequency LOFAR telescope. We also compare our results for different high-precision timing datasets (EPTA+InPTA, PPTA, and LOFAR) of three millisecond pulsars (J0030$+$0451, J1022$+$1001, J2145$-$0450), and find that the solar-wind amplitudes are generally consistent for any individual pulsar, but they can vary from pulsar to pulsar. Finally, we compare our results with those of an independent method on the same LOFAR data of the three millisecond pulsars. We find that differences between the results of the two methods can be mainly attributed to the modelling of dispersion variations in the interstellar medium, rather than the solar wind modelling.

astro-ph.HE↗

arXiv2023

Cell Division and Motility Enable Hexatic Order in Biological Tissues

Biological tissues transform between solid-like and liquid-like states in many fundamental physiological events. Recent experimental observations further suggest that in two-dimensional epithelial tissues these solid-liquid transformations can happen via intermediate states akin to the intermediate hexatic phases observed in equilibrium two-dimensional melting. The hexatic phase is characterized by quasi-long-range (power-law) orientational order but no translational order, thus endowing some structure to an otherwise structureless fluid. While it has been shown that hexatic order in tissue models can be induced by motility and thermal fluctuations, the role of cell division and apoptosis (birth and death) has remained poorly understood, despite its fundamental biological role. Here we study the effect of cell division and apoptosis on global hexatic order within the framework of the self-propelled Voronoi model of tissue. Although cell division naively destroys order and active motility facilitates deformations, we show that their combined action drives a liquid-hexatic-liquid transformation as the motility increases. The hexatic phase is accessed by the delicate balance of dislocation defect generation from cell division and the active binding of disclination-antidisclination pairs from motility. We formulate a mean-field model to elucidate this competition between cell division and motility and the consequent development of hexatic order.

cond-mat.soft↗

arXiv2023

Efficient Meta Neural Heuristic for Multi-Objective Combinatorial Optimization

Recently, neural heuristics based on deep reinforcement learning have exhibited promise in solving multi-objective combinatorial optimization problems (MOCOPs). However, they are still struggling to achieve high learning efficiency and solution quality. To tackle this issue, we propose an efficient meta neural heuristic (EMNH), in which a meta-model is first trained and then fine-tuned with a few steps to solve corresponding single-objective subproblems. Specifically, for the training process, a (partial) architecture-shared multi-task model is leveraged to achieve parallel learning for the meta-model, so as to speed up the training; meanwhile, a scaled symmetric sampling method with respect to the weight vectors is designed to stabilize the training. For the fine-tuning process, an efficient hierarchical method is proposed to systematically tackle all the subproblems. Experimental results on the multi-objective traveling salesman problem (MOTSP), multi-objective capacitated vehicle routing problem (MOCVRP), and multi-objective knapsack problem (MOKP) show that, EMNH is able to outperform the state-of-the-art neural heuristics in terms of solution quality and learning efficiency, and yield competitive solutions to the strong traditional heuristics while consuming much shorter time.

cs.LG↗

arXiv2023

DifFSS: Diffusion Model for Few-Shot Semantic Segmentation

Diffusion models have demonstrated excellent performance in image generation. Although various few-shot semantic segmentation (FSS) models with different network structures have been proposed, performance improvement has reached a bottleneck. This paper presents the first work to leverage the diffusion model for FSS task, called DifFSS. DifFSS, a novel FSS paradigm, can further improve the performance of the state-of-the-art FSS models by a large margin without modifying their network structure. Specifically, we utilize the powerful generation ability of diffusion models to generate diverse auxiliary support images by using the semantic mask, scribble or soft HED boundary of the support image as control conditions. This generation process simulates the variety within the class of the query image, such as color, texture variation, lighting, $etc$. As a result, FSS models can refer to more diverse support images, yielding more robust representations, thereby achieving a consistent improvement in segmentation performance. Extensive experiments on three publicly available datasets based on existing advanced FSS models demonstrate the effectiveness of the diffusion model for FSS task. Furthermore, we explore in detail the impact of different input settings of the diffusion model on segmentation performance. Hopefully, this completely new paradigm will bring inspiration to the study of FSS task integrated with AI-generated content. Code is available at https://github.com/TrinitialChan/DifFSS

cs.CV↗

arXiv2023

Automated Bioinformatics Analysis via AutoBA

With the fast-growing and evolving omics data, the demand for streamlined and adaptable tools to handle the analysis continues to grow. In response to this need, we introduce Auto Bioinformatics Analysis (AutoBA), an autonomous AI agent based on a large language model designed explicitly for conventional omics data analysis. AutoBA simplifies the analytical process by requiring minimal user input while delivering detailed step-by-step plans for various bioinformatics tasks. Through rigorous validation by expert bioinformaticians, AutoBA's robustness and adaptability are affirmed across a diverse range of omics analysis cases, including whole genome sequencing (WGS), RNA sequencing (RNA-seq), single-cell RNA-seq, ChIP-seq, and spatial transcriptomics. AutoBA's unique capacity to self-design analysis processes based on input data variations further underscores its versatility. Compared with online bioinformatic services, AutoBA deploys the analysis locally, preserving data privacy. Moreover, different from the predefined pipeline, AutoBA has adaptability in sync with emerging bioinformatics tools. Overall, AutoBA represents a convenient tool, offering robustness and adaptability for complex omics data analysis.

q-bio.GN↗

arXiv2023

Compressing Vision Transformers for Low-Resource Visual Learning

Vision transformer (ViT) and its variants have swept through visual learning leaderboards and offer state-of-the-art accuracy in tasks such as image classification, object detection, and semantic segmentation by attending to different parts of the visual input and capturing long-range spatial dependencies. However, these models are large and computation-heavy. For instance, the recently proposed ViT-B model has 86M parameters making it impractical for deployment on resource-constrained devices. As a result, their deployment on mobile and edge scenarios is limited. In our work, we aim to take a step toward bringing vision transformers to the edge by utilizing popular model compression techniques such as distillation, pruning, and quantization. Our chosen application environment is an unmanned aerial vehicle (UAV) that is battery-powered and memory-constrained, carrying a single-board computer on the scale of an NVIDIA Jetson Nano with 4GB of RAM. On the other hand, the UAV requires high accuracy close to that of state-of-the-art ViTs to ensure safe object avoidance in autonomous navigation, or correct localization of humans in search-and-rescue. Inference latency should also be minimized given the application requirements. Hence, our target is to enable rapid inference of a vision transformer on an NVIDIA Jetson Nano (4GB) with minimal accuracy loss. This allows us to deploy ViTs on resource-constrained devices, opening up new possibilities in surveillance, environmental monitoring, etc. Our implementation is made available at https://github.com/chensy7/efficient-vit.

cs.CV↗

arXiv2023

DegUIL: Degree-aware Graph Neural Networks for Long-tailed User Identity Linkage

User identity linkage (UIL), matching accounts of a person on different social networks, is a fundamental task in cross-network data mining. Recent works have achieved promising results by exploiting graph neural networks (GNNs) to capture network structure. However, they rarely analyze the realistic node-level bottlenecks that hinder UIL's performance. First, node degrees in a graph vary widely and are long-tailed. A significant fraction of tail nodes with small degrees are underrepresented due to limited structural information, degrading linkage performance seriously. The second bottleneck usually overlooked is super head nodes. It is commonly accepted that head nodes perform well. However, we find that some of them with super high degrees also have difficulty aligning counterparts, due to noise introduced by the randomness of following friends in real-world social graphs. In pursuit of learning ideal representations for these two groups of nodes, this paper proposes a degree-aware model named DegUIL to narrow the degree gap. To this end, our model complements missing neighborhoods for tail nodes and discards redundant structural information for super head nodes in embeddings respectively. Specifically, the neighboring bias is predicted and corrected locally by two modules, which are trained using the knowledge from structurally adequate head nodes. As a result, ideal neighborhoods are obtained for meaningful aggregation in GNNs. Extensive experiments demonstrate the superiority of our model. Our data and code can be found at https://github.com/Longmeix/DegUIL.

cs.SI↗

arXiv2023

ProNet DB: A proteome-wise database for protein surface property representations and RNA-binding profiles

The rapid growth in the number of experimental and predicted protein structures and more complicated protein structures challenge users in computational biology for utilizing the structural information and protein surface property representation. Recently, AlphaFold2 released the comprehensive proteome of various species, and protein surface property representation plays a crucial role in protein-molecule interaction prediction such as protein-protein interaction, protein-nucleic acid interaction, and protein-compound interaction. Here, we proposed the first comprehensive database, namely ProNet DB, which incorporates multiple protein surface representations and RNA-binding landscape for more than 326,175 protein structures covering 16 model organism proteomes from AlphaFold Protein Structure Database (AlphaFold DB) and experimentally validated protein structures deposited in Protein Data Bank (PDB). For each protein, we provided the original protein structure, surface property representation including hydrophobicity, charge distribution, hydrogen bond, interacting face, and RNA-binding landscape such as RNA binding sites and RNA binding preference. To interpret protein surface property representation and RNA binding landscape intuitively, we also integrate Mol* and Online 3D Viewer to visualize the representation on the protein surface. The pre-computed features are available for the users instantaneously and boost computational biology development including molecular mechanism exploration, geometry-based drug discovery and novel therapeutics development. The server is now available on https://proj.cse.cuhk.edu.hk/aihlab/pronet/.

q-bio.QM↗

arXiv2023

Implications of pulsar timing array observations for LISA detections of massive black hole binaries

Pulsar timing arrays (PTAs) and the Laser Interferometer Space Antenna (LISA) will open complementary observational windows on massive black-hole binaries (MBHBs), i.e., with masses in the range $\sim 10^6 - 10^{10}\,$ M$_{\odot}$. While PTAs may detect a stochastic gravitational-wave background from a population of MBHBs, during operation LISA will detect individual merging MBHBs. To demonstrate the profound interplay between LISA and PTAs, we estimate the number of MBHB mergers that one can expect to observe with LISA by extrapolating direct observational constraints on the MBHB merger rate inferred from PTA data. For this, we postulate that the common signal observed by PTAs (and consistent with the increased evidence recently reported) is an astrophysical background sourced by a single MBHB population. We then constrain the LISA detection rate, $\mathcal{R}$, in the mass-redshift space by combining our Bayesian-inferred merger rate with LISA's sensitivity to spin-aligned, inspiral-merger-ringdown waveforms. Using an astrophysically-informed formation model, we predict a 95$\%$ upper limit on the detection rate of $\mathcal{R} < 134\,{\rm yr}^{-1}$ for binaries with total masses in the range $10^7 - 10^8\,$ M$_{\odot}$. For higher masses, i.e., $>10^8\,$ M$_{\odot}$, we find $\mathcal{R} < 2\,(1)\,\mathrm{yr}^{-1}$ using an astrophysically-informed (agnostic) formation model, rising to $11\,(6)\,\mathrm{yr}^{-1}$ if the LISA sensitivity bandwidth extends down to $10^{-5}$ Hz. Forecasts of LISA science potential with PTA background measurements should improve as PTAs continue their search.

astro-ph.HE↗