Search

Your search keyword '"Cui, Xingyu"' showing total 15 results
Start Over Author "Cui, Xingyu" Search Limiters Full Text
15 results on '"Cui, Xingyu"'

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

Author

Wei, Xiang, Chen, Yufeng, Cheng, Ning, Cui, Xingyu, Xu, Jinan, and Han, Wenjuan

Subjects
Computer Science - Computation and Language
Abstract

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.

Published
2023

2. ChatIE: Zero-Shot Information Extraction via Chatting with ChatGPT

Author

Wei, Xiang, Cui, Xingyu, Cheng, Ning, Wang, Xiaobin, Zhang, Xin, Huang, Shen, Xie, Pengjun, Xu, Jinan, Chen, Yufeng, Zhang, Meishan, Jiang, Yong, and Han, Wenjuan

Subjects
Computer Science - Computation and Language
Abstract

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.

Published
2023

6. Zero-Shot Information Extraction via Chatting with ChatGPT

Author

Wei, Xiang, Cui, Xingyu, Cheng, Ning, Wang, Xiaobin, Zhang, Xin, Huang, Shen, Xie, Pengjun, Xu, Jinan, Chen, Yufeng, Zhang, Meishan, Jiang, Yong, and Han, Wenjuan

Subjects
FOS: Computer and information sciences, Computer Science - Computation and Language, Computation and Language (cs.CL)
Abstract

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.

Published
2023

8. Fractal Superconducting Nanowire Single-Photon Detectors and Their Applications in Imaging

Author

Feng, Yifan, Meng, Yun, Zou, Kai, Hu, Nan, Hao, Zifan, Cui, Xingyu, Yin, Xiangjun, Yang, Jingyu, Gyger, Samuel, Steinhauer, Stephan, Zwiller, Val, Hu, Xiaolong, Feng, Yifan, Meng, Yun, Zou, Kai, Hu, Nan, Hao, Zifan, Cui, Xingyu, Yin, Xiangjun, Yang, Jingyu, Gyger, Samuel, Steinhauer, Stephan, Zwiller, Val, and Hu, Xiaolong

Abstract

We present our research on fractal superconducting nanowire single-photon detectors and their applications in light detection and ranging (LiDAR), full-Stokes polarimetric imaging, and non-line-of-sight imaging., QC 20230831

Published
2022
Full Text
View/download PDF

10. Co2N Nanoparticles Anchored on N‐Doped Active Carbon as Catalyst for Oxygen Reduction Reaction in Zinc–Air Battery.

Author

Wu, Xianli, Han, Guosheng, Wen, Hao, Liu, Yanyan, Han, Lei, Cui, Xingyu, Kou, Jiajing, Li, Baojun, and Jiang, Jianchun

Subjects
OXYGEN reduction, DOPING agents (Chemistry), OXYGEN electrodes, DENSITY functional theory, FERMI level
Abstract

The development of efficient catalytic electrode toward oxygen reduction reaction (ORR) is still a great challenge for the wide use of zinc–air batteries. Herein, Co2N nanoparticles (NPs) anchored on N‐doped carbon from cattail were verified with excellent catalytic performances for ORR. The onset and half‐wave potentials over the optimal catalyst reach to 0.96 V and 0.84 V, respectively. Current retention rates of 96.8% after 22‐h test and 98.8% after running 1600 s were obtained in 1 M methanol solution. Density functional theory simulation proposes an apparently increased electronic states of Co2N in N‐doped carbon layer close to the Fermi level. Higher charge density, favorable adsorption, and charge transfer of intermediates originate from the coexistence of Co2N NPs and N atoms in carbon skeleton. The superior catalytic activity of composites also was confirmed in zinc–air batteries. This novel catalytic property and controllable preparation approach of Co2N‐carbon composites provide a promising avenue to fabricate metal‐containing catalytically active carbon from biomass. [ABSTRACT FROM AUTHOR]

Published
2022
Full Text
View/download PDF

15. Effect of Polyacrylic Acid Addition on Rheology of SiC- Al2 O3- ZrO2(3 Y) Mixed Suspensions.

Author

Zhang, Ning, Liang, Bin, Zhao, Xiaojun, Cui, Xingyu, Kan, Hongmin, and Liu, Hao

Subjects
POLYACRYLIC acid, ADDITION reactions, RHEOLOGY, SILICON carbide, ALUMINUM oxide, ZIRCONIUM oxide, SUSPENSIONS (Chemistry), SLIP casting, GELCASTING
Abstract

A suspension with good rheology and high stability is crucial for slip casting and gelcasting technology. However, a mixed suspension from two or more different powders usually has bad rheology because of the easy agglomeration of mixed powders caused by the attractive force between the powders with heterocharges. We studied the surface modification of the each single-component powders ( SiC, Al2O3, ZrO2(3 Y) powders) and the SiC- Al2O3- ZrO2(3 Y) mixed powders to increase the repulsive force by adjusting the pH value and adding polyacrylic acid ( PAA) as dispersant. The PAA addition effects on the SiC- Al2O3- ZrO2(3 Y) mixture were investigated in terms of zeta potential, pH range for heterocharge region, dispersion of the mixed powders and rheology of the mixed slurry based on the study of each unary suspensions. The results show that before surface modification the SiC- Al2O3- ZrO2(3 Y) mixed powders were agglomerated severely because they were in the heterocharge region with a broad pH range from 3.5 to 8.25, while after surface modification ( pH = 10.5, PAA = 0.8wt%) the heterocharge region was narrowed with a relatively narrower pH range from 2.6 to 3.7. The mixed powders with homocharges were dispersed well because of the great electrostatic repulsive force and steric hindrance offered by PAA and the mixed suspensions had favorable rheology. [ABSTRACT FROM AUTHOR]

Published
2013
Full Text
View/download PDF

Catalog

Books, media, physical & digital resources
See catalog results