arXiv subjects

Xiang Shi

Publications and source records attributed to Xiang Shi.

At least 37 records · Page 2Linked to original sources

arXiv2024

Every Part Matters: Integrity Verification of Scientific Figures Based on Multimodal Large Language Models

This paper tackles a key issue in the interpretation of scientific figures: the fine-grained alignment of text and figures. It advances beyond prior research that primarily dealt with straightforward, data-driven visualizations such as bar and pie charts and only offered a basic understanding of diagrams through captioning and classification. We introduce a novel task, Figure Integrity Verification, designed to evaluate the precision of technologies in aligning textual knowledge with visual elements in scientific figures. To support this, we develop a semi-automated method for constructing a large-scale dataset, Figure-seg, specifically designed for this task. Additionally, we propose an innovative framework, Every Part Matters (EPM), which leverages Multimodal Large Language Models (MLLMs) to not only incrementally improve the alignment and verification of text-figure integrity but also enhance integrity through analogical reasoning. Our comprehensive experiments show that these innovations substantially improve upon existing methods, allowing for more precise and thorough analysis of complex scientific figures. This progress not only enhances our understanding of multimodal technologies but also stimulates further research and practical applications across fields requiring the accurate interpretation of complex visual data.

cs.CL↗

arXiv2024

Let's Learn Step by Step: Enhancing In-Context Learning Ability with Curriculum Learning

Demonstration ordering, which is an important strategy for in-context learning (ICL), can significantly affects the performance of large language models (LLMs). However, most of the current approaches of ordering require high computational costs to introduce the priori knowledge. In this paper, inspired by the human learning process, we propose a simple but effective demonstration ordering method for ICL, named the few-shot In-Context Curriculum Learning (ICCL). The ICCL implies gradually increasing the complexity of prompt demonstrations during the inference process. The difficulty can be assessed by human experts or LLMs-driven metrics, such as perplexity. Then we design extensive experiments to discuss the effectiveness of the ICCL at both corpus-level and instance-level. Moreover, we also investigate the formation mechanism of LLM's ICCL capability. Experimental results demonstrate that ICCL, developed during the instruction-tuning stage, is effective for representative open-source LLMs. To facilitate further research and applications by other scholars, we make the code publicly available.

cs.CL↗

arXiv2023

Atom: Neural Traffic Compression with Spatio-Temporal Graph Neural Networks

Storing network traffic data is key to efficient network management; however, it is becoming more challenging and costly due to the ever-increasing data transmission rates, traffic volumes, and connected devices. In this paper, we explore the use of neural architectures for network traffic compression. Specifically, we consider a network scenario with multiple measurement points in a network topology. Such measurements can be interpreted as multiple time series that exhibit spatial and temporal correlations induced by network topology, routing, or user behavior. We present \textit{Atom}, a neural traffic compression method that leverages spatial and temporal correlations present in network traffic. \textit{Atom} implements a customized spatio-temporal graph neural network design that effectively exploits both types of correlations simultaneously. The experimental results show that \textit{Atom} can outperform GZIP's compression ratios by 50\%-65\% on three real-world networks.

cs.NI↗

arXiv2023

Know Where to Go: Make LLM a Relevant, Responsible, and Trustworthy Searcher

The advent of Large Language Models (LLMs) has shown the potential to improve relevance and provide direct answers in web searches. However, challenges arise in validating the reliability of generated results and the credibility of contributing sources, due to the limitations of traditional information retrieval algorithms and the LLM hallucination problem. Aiming to create a "PageRank" for the LLM era, we strive to transform LLM into a relevant, responsible, and trustworthy searcher. We propose a novel generative retrieval framework leveraging the knowledge of LLMs to foster a direct link between queries and online sources. This framework consists of three core modules: Generator, Validator, and Optimizer, each focusing on generating trustworthy online sources, verifying source reliability, and refining unreliable sources, respectively. Extensive experiments and evaluations highlight our method's superior relevance, responsibility, and trustfulness against various SOTA methods.

cs.IR↗

arXiv2023

RouteNet-Fermi: Network Modeling with Graph Neural Networks

Network models are an essential block of modern networks. For example, they are widely used in network planning and optimization. However, as networks increase in scale and complexity, some models present limitations, such as the assumption of Markovian traffic in queuing theory models, or the high computational cost of network simulators. Recent advances in machine learning, such as Graph Neural Networks (GNN), are enabling a new generation of network models that are data-driven and can learn complex non-linear behaviors. In this paper, we present RouteNet-Fermi, a custom GNN model that shares the same goals as Queuing Theory, while being considerably more accurate in the presence of realistic traffic models. The proposed model predicts accurately the delay, jitter, and packet loss of a network. We have tested RouteNet-Fermi in networks of increasing size (up to 300 nodes), including samples with mixed traffic profiles -- e.g., with complex non-Markovian models -- and arbitrary routing and queue scheduling configurations. Our experimental results show that RouteNet-Fermi achieves similar accuracy as computationally-expensive packet-level simulators and scales accurately to larger networks. Our model produces delay estimates with a mean relative error of 6.24% when applied to a test dataset of 1,000 samples, including network topologies one order of magnitude larger than those seen during training. Finally, we have also evaluated RouteNet-Fermi with measurements from a physical testbed and packet traces from a real-life network.

cs.NI↗

arXiv2023

GraphCC: A Practical Graph Learning-based Approach to Congestion Control in Datacenters

Congestion Control (CC) plays a fundamental role in optimizing traffic in Data Center Networks (DCN). Currently, DCNs mainly implement two main CC protocols: DCTCP and DCQCN. Both protocols -- and their main variants -- are based on Explicit Congestion Notification (ECN), where intermediate switches mark packets when they detect congestion. The ECN configuration is thus a crucial aspect on the performance of CC protocols. Nowadays, network experts set static ECN parameters carefully selected to optimize the average network performance. However, today's high-speed DCNs experience quick and abrupt changes that severely change the network state (e.g., dynamic traffic workloads, incast events, failures). This leads to under-utilization and sub-optimal performance. This paper presents GraphCC, a novel Machine Learning-based framework for in-network CC optimization. Our distributed solution relies on a novel combination of Multi-agent Reinforcement Learning (MARL) and Graph Neural Networks (GNN), and it is compatible with widely deployed ECN-based CC protocols. GraphCC deploys distributed agents on switches that communicate with their neighbors to cooperate and optimize the global ECN configuration. In our evaluation, we test the performance of GraphCC under a wide variety of scenarios, focusing on the capability of this solution to adapt to new scenarios unseen during training (e.g., new traffic workloads, failures, upgrades). We compare GraphCC with a state-of-the-art MARL-based solution for ECN tuning -- ACC -- and observe that our proposed solution outperforms the state-of-the-art baseline in all of the evaluation scenarios, showing improvements up to $20\%$ in Flow Completion Time as well as significant reductions in buffer occupancy ($38.0-85.7\%$).

cs.NI↗

arXiv2023

MAGNNETO: A Graph Neural Network-based Multi-Agent system for Traffic Engineering

Current trends in networking propose the use of Machine Learning (ML) for a wide variety of network optimization tasks. As such, many efforts have been made to produce ML-based solutions for Traffic Engineering (TE), which is a fundamental problem in ISP networks. Nowadays, state-of-the-art TE optimizers rely on traditional optimization techniques, such as Local search, Constraint Programming, or Linear programming. In this paper, we present MAGNNETO, a distributed ML-based framework that leverages Multi-Agent Reinforcement Learning and Graph Neural Networks for distributed TE optimization. MAGNNETO deploys a set of agents across the network that learn and communicate in a distributed fashion via message exchanges between neighboring agents. Particularly, we apply this framework to optimize link weights in OSPF, with the goal of minimizing network congestion. In our evaluation, we compare MAGNNETO against several state-of-the-art TE optimizers in more than 75 topologies (up to 153 nodes and 354 links), including realistic traffic loads. Our experimental results show that, thanks to its distributed nature, MAGNNETO achieves comparable performance to state-of-the-art TE optimizers with significantly lower execution times. Moreover, our ML-based solution demonstrates a strong generalization capability to successfully operate in new networks unseen during training.

cs.NI↗

arXiv2023

Leveraging Spatial and Temporal Correlations for Network Traffic Compression

The deployment of modern network applications is increasing the network size and traffic volumes at an unprecedented pace. Storing network-related information (e.g., traffic traces) is key to enable efficient network management. However, this task is becoming more challenging due to the ever-increasing data transmission rates and traffic volumes. In this paper, we present a novel method for network traffic compression that exploits spatial and temporal patterns naturally present in network traffic. We consider a realistic scenario where traffic measurements are performed at multiple links of a network topology using tools like SNMP or NetFlow. Such measurements can be seen as multiple time series that exhibit spatial and temporal correlations induced by the network topology, routing or user behavior. Our method leverages graph learning methods to effectively exploit both types of correlations for traffic compression. The experimental results show that our solution is able to outperform GZIP, the \textit{de facto} traffic compression method, improving by 50\%-65\% the compression ratio on three real-world networks.

cs.NI↗

arXiv2022

Scattered Light in STIS Grating G230LB

The G230LB grating used with STIS's CCD detector scatters red light. In red objects, the scattered light mingles with the ultraviolet signal, causing spurious short-wavelength flux and weakening absorption features. Recent calibration observations characterize the scattered light using duplicate observations with the MAMA detector and similar grating G230L. The full two-dimensional spectrum contains little helpful information to mitigate the scattered light problem. For one-dimensional, extracted spectra, the scattered light can be approximately modeled as a ramped pedestal whose amplitude is proportional to the object's V-band flux. We present formulae for scattered light corrections. For stars warmer than G0 spectral type, correction is superfluous. Off-slit-center positioning appears not to affect the properties of the scattered light. Therefore, we are able to extrapolate correction formulae for extended objects from the point source formulae. Polynomials for flux corrections due to off-center slit positioning in the 0.2 arcsec slit are also tabulated.

astro-ph.IM↗

arXiv2022

Chemically Peculiar Stars in Integrated Light Stellar Population Models and Local Group Galaxies

Integrated-light models that incorporate common types of chemically peculiar (CP) stars are assembled using synthetic spectra. Selected spectral features encode significant age information for populations with ages $\sim$50 Myr $<$ age $< \sim$2 Gyr. Due to the alleviation of template mismatch, the inclusion of CP star features in model spectra improves the accuracy of recovered stellar population parameters, but we are not able to show that new or unique age information can be extracted from the weak CP features compared to continuum fitting and strong-feature strengths, at least at the present state of the art. An age-extraction routine that recovers 2- and 3-burst age structures is employed to analyze the spectra of local group galaxies. NGC 224 (M31) has a stellar population too old for the types of CP stars we examine. NGC 221 (M32) also shows no CP spectral features. It appears to contain a component at age $\sim$1 Gyr at 1% by mass in addition to its dominant 4.7 Gyr population. Unlike SDSS galaxy spectrum averages, NGC 205 (M110) contains no features due to HgMn stars. This excludes the age range associated with HgMn production, and its near-nuclear spectrum is best fit by a 68$\pm$2 Myr population superimposed on an older population with a 1.85$\pm$0.1 Gyr component. Both NGC 205 and NGC 221 have an ancient component whose mass is not easy to constrain given the overwhelming light-dominance of the younger populations.

astro-ph.GA↗

arXiv2022

ENERO: Efficient Real-Time WAN Routing Optimization with Deep Reinforcement Learning

Wide Area Networks (WAN) are a key infrastructure in today's society. During the last years, WANs have seen a considerable increase in network's traffic and network applications, imposing new requirements on existing network technologies (e.g., low latency and high throughput). Consequently, Internet Service Providers (ISP) are under pressure to ensure the customer's Quality of Service and fulfill Service Level Agreements. Network operators leverage Traffic Engineering (TE) techniques to efficiently manage network's resources. However, WAN's traffic can drastically change during time and the connectivity can be affected due to external factors (e.g., link failures). Therefore, TE solutions must be able to adapt to dynamic scenarios in real-time. In this paper we propose Enero, an efficient real-time TE solution based on a two-stage optimization process. In the first one, Enero leverages Deep Reinforcement Learning (DRL) to optimize the routing configuration by generating a long-term TE strategy. To enable efficient operation over dynamic network scenarios (e.g., when link failures occur), we integrated a Graph Neural Network into the DRL agent. In the second stage, Enero uses a Local Search algorithm to improve DRL's solution without adding computational overhead to the optimization process. The experimental results indicate that Enero is able to operate in real-world dynamic network topologies in 4.5 seconds on average for topologies up to 100 edges.

cs.NI↗

arXiv2022

Network Digital Twin: Context, Enabling Technologies and Opportunities

The proliferation of emergent network applications (e.g., telesurgery, metaverse) is increasing the difficulty of managing modern communication networks. These applications entail stringent network requirements (e.g., ultra-low deterministic latency), which hinders network operators to manage their resources efficiently. In this article, we introduce the network digital twin (NDT), a renovated concept of classical network modeling tools whose goal is to build accurate data-driven network models that can operate in real-time. We describe the general architecture of the NDT and argue that modern machine learning (ML) technologies enable building some of its core components. Then, we present a case study that leverages a ML-based NDT for network performance evaluation and apply it to routing optimization in a QoS-aware use case. Lastly, we describe some key open challenges and research opportunities yet to be explored to achieve effective deployment of NDTs in real-world networks.

cs.NI↗

arXiv2022

Graph Neural Networks for Communication Networks: Context, Use Cases and Opportunities

Graph neural networks (GNN) have shown outstanding applications in many fields where data is fundamentally represented as graphs (e.g., chemistry, biology, recommendation systems). In this vein, communication networks comprise many fundamental components that are naturally represented in a graph-structured manner (e.g., topology, configurations, traffic flows). This position article presents GNNs as a fundamental tool for modeling, control and management of communication networks. GNNs represent a new generation of data-driven models that can accurately learn and reproduce the complex behaviors behind real networks. As a result, such models can be applied to a wide variety of networking use cases, such as planning, online optimization, or troubleshooting. The main advantage of GNNs over traditional neural networks lies in its unprecedented generalization capabilities when applied to other networks and configurations unseen during training, which is a critical feature for achieving practical data-driven solutions for networking. This article comprises a brief tutorial on GNNs and their possible applications to communication networks. To showcase the potential of this technology, we present two use cases with state-of-the-art GNN models respectively applied to wired and wireless networks. Lastly, we delve into the key open challenges and opportunities yet to be explored in this novel research area.

cs.NI↗

arXiv2022

Accelerating Deep Reinforcement Learning for Digital Twin Network Optimization with Evolutionary Strategies

The recent growth of emergent network applications (e.g., satellite networks, vehicular networks) is increasing the complexity of managing modern communication networks. As a result, the community proposed the Digital Twin Networks (DTN) as a key enabler of efficient network management. Network operators can leverage the DTN to perform different optimization tasks (e.g., Traffic Engineering, Network Planning). Deep Reinforcement Learning (DRL) showed a high performance when applied to solve network optimization problems. In the context of DTN, DRL can be leveraged to solve optimization problems without directly impacting the real-world network behavior. However, DRL scales poorly with the problem size and complexity. In this paper, we explore the use of Evolutionary Strategies (ES) to train DRL agents for solving a routing optimization problem. The experimental results show that ES achieved a training time speed-up of 128 and 6 for the NSFNET and GEANT2 topologies respectively.

cs.NI↗

arXiv2022

Individual element sensitivity for stellar evolutionary isochrones

Stellar evolution calculations with variable abundance ratios were used to gauge the effects on temperatures, luminosities, and lifetimes in various phases. Individual elements C, N, O, Mg, Si, and Fe were included. Most of the effect relevant to integrated light models is contained in the temperature variable, as opposed to the timescale or luminosity. We derive a recipe for including abundance-sensitive temperature effects that is applicable to existing isochrone grids. The resultant enhanced isochrones are incorporated into composite stellar population models and compared with galaxy data from the Sloan Digital Sky Survey. A severe oxygen-age degeneracy is apparent, 2 - 3 Gyr per 0.1 dex in [O/R], where R represents a heavy element such as Fe. Over the range of early-type galaxy velocity dispersion, the spans of all abundance ratios are reduced but the age range increases, systematically older. Allowing Fe-peak elements the freedom to vary accentuates this increase of age span. Overall, these results sharpen the age-mass correlation known as downsizing but decrease the steepness of abundance ratio gradients. Both of these observations, in turn, imply a more robust contribution from gas free mergers in the histories of typical elliptical galaxies.

astro-ph.GA↗

arXiv2022

Digital Twin Network: Opportunities and Challenges

The proliferation of emergent network applications (e.g., AR/VR, telesurgery, real-time communications) is increasing the difficulty of managing modern communication networks. These applications typically have stringent requirements (e.g., ultra-low deterministic latency), making it more difficult for network operators to manage their network resources efficiently. In this article, we propose the Digital Twin Network (DTN) as a key enabler for efficient network management in modern networks. We describe the general architecture of the DTN and argue that recent trends in Machine Learning (ML) enable building a DTN that efficiently and accurately mimics real-world networks. In addition, we explore the main ML technologies that enable developing the components of the DTN architecture. Finally, we describe the open challenges that the research community has to address in the upcoming years in order to enable the deployment of the DTN in real-world scenarios.

cs.NI↗

arXiv2021

Improving Text-to-SQL with Schema Dependency Learning

Text-to-SQL aims to map natural language questions to SQL queries. The sketch-based method combined with execution-guided (EG) decoding strategy has shown a strong performance on the WikiSQL benchmark. However, execution-guided decoding relies on database execution, which significantly slows down the inference process and is hence unsatisfactory for many real-world applications. In this paper, we present the Schema Dependency guided multi-task Text-to-SQL model (SDSQL) to guide the network to effectively capture the interactions between questions and schemas. The proposed model outperforms all existing methods in both the settings with or without EG. We show the schema dependency learning partially cover the benefit from EG and alleviates the need for it. SDSQL without EG significantly reduces time consumption during inference, sacrificing only a small amount of performance and provides more flexibility for downstream applications.

cs.CL↗

arXiv2018

Spatial statistics of atmospheric particulate matter in China

In this paper, the spatial dynamics of the atmospheric particulate matters (resp. PM$_{10}$ and PM$_{2.5}$) are studied using turbulence methodologies. It is found experimentally that the spatial correlation function $ρ(r)$ shows a log-law on the mesoscale range, i.e., $50\le r\le 500\,{km}$, with an experimental scaling exponent $β=0.45$. The spatial structure function shows a power-law behavior on the mesoscale range $90\le r\le 500\,{km}$. The experimental scaling exponent $ζ(q)$ is convex, showing that the intermittent correction is relevant in characterizing the spatial dynamic of particulate matter. The measured singularity spectrum $f(α)$ also shows its multifractal nature. Experimentally, the particulate matter is more intermittent than the passive scalar, which could be partially due to the \red{mesoscale movements} of the atmosphere, and also due to local sources, such as local industry activities.

physics.ao-ph↗