arXiv subjects

Xiang Wei

Publications and source records attributed to Xiang Wei.

28 records · Page 2Linked to original sources

arXiv2024

ChatIE: Zero-Shot Information Extraction via Chatting with ChatGPT

Zero-shot information extraction (IE) aims to build IE systems from the unannotated text. It is challenging due to involving little human intervention. Challenging but worthwhile, zero-shot IE reduces the time and effort that data labeling takes. Recent efforts on large language models (LLMs, e.g., GPT-3, ChatGPT) show promising performance on zero-shot settings, thus inspiring us to explore prompt-based methods. In this work, we ask whether strong IE models can be constructed by directly prompting LLMs. Specifically, we transform the zero-shot IE task into a multi-turn question-answering problem with a two-stage framework (ChatIE). With the power of ChatGPT, we extensively evaluate our framework on three IE tasks: entity-relation triple extract, named entity recognition, and event extraction. Empirical results on six datasets across two languages show that ChatIE achieves impressive performance and even surpasses some full-shot models on several datasets (e.g., NYT11-HRL). We believe that our work could shed light on building IE models with limited resources.

cs.CL↗

arXiv2024

TransGPT: Multi-modal Generative Pre-trained Transformer for Transportation

Natural language processing (NLP) is a key component of intelligent transportation systems (ITS), but it faces many challenges in the transportation domain, such as domain-specific knowledge and data, and multi-modal inputs and outputs. This paper presents TransGPT, a novel (multi-modal) large language model for the transportation domain, which consists of two independent variants: TransGPT-SM for single-modal data and TransGPT-MM for multi-modal data. TransGPT-SM is finetuned on a single-modal Transportation dataset (STD) that contains textual data from various sources in the transportation domain. TransGPT-MM is finetuned on a multi-modal Transportation dataset (MTD) that we manually collected from three areas of the transportation domain: driving tests, traffic signs, and landmarks. We evaluate TransGPT on several benchmark datasets for different tasks in the transportation domain, and show that it outperforms baseline models on most tasks. We also showcase the potential applications of TransGPT for traffic analysis and modeling, such as generating synthetic traffic scenarios, explaining traffic phenomena, answering traffic-related questions, providing traffic recommendations, and generating traffic reports. This work advances the state-of-the-art of NLP in the transportation domain and provides a useful tool for ITS researchers and practitioners.

cs.CL↗

arXiv2023

Levenshtein Distance Embedding with Poisson Regression for DNA Storage

Efficient computation or approximation of Levenshtein distance, a widely-used metric for evaluating sequence similarity, has attracted significant attention with the emergence of DNA storage and other biological applications. Sequence embedding, which maps Levenshtein distance to a conventional distance between embedding vectors, has emerged as a promising solution. In this paper, a novel neural network-based sequence embedding technique using Poisson regression is proposed. We first provide a theoretical analysis of the impact of embedding dimension on model performance and present a criterion for selecting an appropriate embedding dimension. Under this embedding dimension, the Poisson regression is introduced by assuming the Levenshtein distance between sequences of fixed length following a Poisson distribution, which naturally aligns with the definition of Levenshtein distance. Moreover, from the perspective of the distribution of embedding distances, Poisson regression approximates the negative log likelihood of the chi-squared distribution and offers advancements in removing the skewness. Through comprehensive experiments on real DNA storage data, we demonstrate the superior performance of the proposed method compared to state-of-the-art approaches.

cs.LG↗

arXiv2023

CollabKG: A Learnable Human-Machine-Cooperative Information Extraction Toolkit for (Event) Knowledge Graph Construction

In order to construct or extend entity-centric and event-centric knowledge graphs (KG and EKG), the information extraction (IE) annotation toolkit is essential. However, existing IE toolkits have several non-trivial problems, such as not supporting multi-tasks, not supporting automatic updates. In this work, we present CollabKG, a learnable human-machine-cooperative IE toolkit for KG and EKG construction. Specifically, for the multi-task issue, CollabKG unifies different IE subtasks, including named entity recognition (NER), entity-relation triple extraction (RE), and event extraction (EE), and supports both KG and EKG. Then, combining advanced prompting-based IE technology, the human-machine-cooperation mechanism with LLMs as the assistant machine is presented which can provide a lower cost as well as a higher performance. Lastly, owing to the two-way interaction between the human and machine, CollabKG with learning ability allows self-renewal. Besides, CollabKG has several appealing features (e.g., customization, training-free, propagation, etc.) that make the system powerful, easy-to-use, and high-productivity. We holistically compare our toolkit with other existing tools on these features. Human evaluation quantitatively illustrates that CollabKG significantly improves annotation quality, efficiency, and stability simultaneously.

cs.CL↗

arXiv2023

Prompt What You Need: Enhancing Segmentation in Rainy Scenes with Anchor-based Prompting

Semantic segmentation in rainy scenes is a challenging task due to the complex environment, class distribution imbalance, and limited annotated data. To address these challenges, we propose a novel framework that utilizes semi-supervised learning and pre-trained segmentation foundation model to achieve superior performance. Specifically, our framework leverages the semi-supervised model as the basis for generating raw semantic segmentation results, while also serving as a guiding force to prompt pre-trained foundation model to compensate for knowledge gaps with entropy-based anchors. In addition, to minimize the impact of irrelevant segmentation masks generated by the pre-trained foundation model, we also propose a mask filtering and fusion mechanism that optimizes raw semantic segmentation results based on the principle of minimum risk. The proposed framework achieves superior segmentation performance on the Rainy WCity dataset and is awarded the first prize in the sub-track of STRAIN in ICME 2023 Grand Challenges.

cs.CV↗

arXiv2022

Faddeev fixed-center approximation to the $D\bar{D}K$ system and the hidden charm $K_{c\bar{c}}(4180)$ state

We perform a theoretical study on the $D\bar{D}K$ three body system, using the fixed center approximation to the Faddeev equations, considering the interaction between $D$ and $K$, $D$ and $\bar{D}$ from the chiral unitary approach. We assume the scattering of $K$ meson on a clusterized system $D\bar{D}$, where a scalar meson $X(3720)$ could be formed. Thanks to the strong $DK$ interaction, where the scalar $D^*_{s0}(2317)$ meson is dynamically generated, a resonance structure shows up in the modulus squared of the three body $K$-$(D\bar{D})_{X(3720)}$ scattering amplitude and supports that a $D\bar{D}K$ bound state can be formed. The result is in agreement with previous theoretical studies, which claim a new excited hidden charm $K$ meson, $K_{c\bar{c}}(4180)$ with quantum numbers $I(J^P) = \frac{1}{2}(0^-)$ and mass about $4180$ MeV. It is expected that these theoretical results motivate its search in experimental measurements.

hep-ph↗

arXiv2020

High-resolution wide-field OCT angiography with a self-navigation method to correct microsaccades and blinks

In this study, we demonstrate a novel self-navigated motion correction method that suppresses eye motion and blinking artifacts on wide-field optical coherence tomographic angiography (OCTA) without requiring any hardware modification. Highly efficient GPU-based, real-time OCTA image acquisition and processing software was developed to detect eye motion artifacts. The algorithm includes an instantaneous motion index that evaluates the strength of motion artifact on en face OCTA images. Areas with suprathreshold motion and eye blinking artifacts are automatically rescanned in real-time. Both healthy eyes and eyes with diabetic retinopathy were imaged, and the self-navigated motion correction performance was demonstrated.

physics.med-ph↗

arXiv2018

Improving the Improved Training of Wasserstein GANs: A Consistency Term and Its Dual Effect

Despite being impactful on a variety of problems and applications, the generative adversarial nets (GANs) are remarkably difficult to train. This issue is formally analyzed by \cite{arjovsky2017towards}, who also propose an alternative direction to avoid the caveats in the minmax two-player training of GANs. The corresponding algorithm, called Wasserstein GAN (WGAN), hinges on the 1-Lipschitz continuity of the discriminator. In this paper, we propose a novel approach to enforcing the Lipschitz continuity in the training procedure of WGANs. Our approach seamlessly connects WGAN with one of the recent semi-supervised learning methods. As a result, it gives rise to not only better photo-realistic samples than the previous methods but also state-of-the-art semi-supervised learning results. In particular, our approach gives rise to the inception score of more than 5.0 with only 1,000 CIFAR-10 images and is the first that exceeds the accuracy of 90% on the CIFAR-10 dataset using only 4,000 labeled images, to the best of our knowledge.

cs.CV↗

arXiv2017

Cooperative epidemic spreading on a two-layered interconnected network

This study is concerned with the dynamical behaviors of epidemic spreading over a two-layered interconnected network. Three models in different levels are proposed to describe cooperative spreading processes over the interconnected network, wherein the disease in one network can spread to the other. Theoretical analysis is provided for each model to reveal that the global epidemic threshold in the interconnected network is not larger than the epidemic thresholds for the two isolated layered networks. In particular, in an interconnected homogenous network, detailed theoretical analysis is presented, which allows quick and accurate calculations of the global epidemic threshold. Moreover, in an interconnected heterogeneous network with inter-layer correlation between node degrees, it is found that the inter-layer correlation coefficient has little impact on the epidemic threshold, but has significant impact on the total prevalence. Simulations further verify the analytical results, showing that cooperative epidemic processes promote the spreading of diseases.

physics.soc-ph↗

arXiv2015

Counterpart synchronization of duplex networks with delayed nodes and noise perturbation

In the real world, many complex systems are represented not by single networks but rather by sets of interdependent ones. In these specific networks, nodes in one network mutually interact with nodes in other networks. This paper focuses on a simple representative case of two-layer networks (the so-called duplex networks) with unidirectional inter-layer couplings. That is, each node in one network depends on a counterpart in the other network. Accordingly, the former network is called the response layer and the latter network is the drive layer. Specifically, synchronization between each node in the drive layer and its counterpart in the response layer (counterpart synchronization, or CS) of this sort of duplex networks with delayed nodes and noise perturbation is investigated. Based on the LaSalle-type invariance principle, a control technique is proposed and a sufficient condition is developed for realizing counterpart synchronization of duplex networks. Furthermore, two corollaries are derived as special cases. In addition, node dynamics within each layer can be various and topologies of the two layers are not necessarily identical. Therefore, the proposed synchronization method can be applied to a wide range of multiplex networks. Numerical examples are provided to illustrate the feasibility and effectiveness of the results.

nlin.CD↗