Prompt2Gaussia: Uncertain Prompt-learning for Script Event Prediction

no code implementations • 4 Aug 2023 • Shiyao Cui, Xin Cong, Jiawei Sheng, Xuebin Wang, Tingwen Liu, Jinqiao Shi

In this paper, we regard public pre-trained language models as knowledge bases and automatically mine the script-related knowledge via prompt-learning.

Enhancing Multimodal Entity and Relation Extraction with Variational Information Bottleneck

no code implementations • 5 Apr 2023 • Shiyao Cui, Jiangxia Cao, Xin Cong, Jiawei Sheng, Quangang Li, Tingwen Liu, Jinqiao Shi

For the first issue, a refinement-regularizer probes the information-bottleneck principle to balance the predictive evidence and noisy information, yielding expressive representations for prediction.

named-entity-recognition Named Entity Recognition +3

URM4DMU: an user represention model for darknet markets users

no code implementations • 19 Mar 2023 • Hongmeng Liu, Jiapeng Zhao, Yixuan Huo, Yuyan Wang, Chun Liao, Liyan Shen, Shiyao Cui, Jinqiao Shi

Traditional user representation methods mainly rely on modeling the text information of posts and cannot capture the temporal content and the forum interaction of posts.

The style transformer with common knowledge optimization for image-text retrieval

no code implementations • 1 Mar 2023 • Wenrui Li, Zhengyu Ma, Jinqiao Shi, Xiaopeng Fan

The main module is the common knowledge adaptor (CKA) with both the style embedding extractor (SEE) and the common knowledge optimization (CKO) modules.

Retrieval Text Retrieval

Event Causality Extraction with Event Argument Correlations

1 code implementation • COLING 2022 • Shiyao Cui, Jiawei Sheng, Xin Cong, Quangang Li, Tingwen Liu, Jinqiao Shi

Event Causality Identification (ECI), which aims to detect whether a causality relation exists between two given textual events, is an important task for event causality understanding.

Event Causality Identification

PrUE: Distilling Knowledge from Sparse Teacher Networks

1 code implementation • 3 Jul 2022 • Shaopu Wang, Xiaojun Chen, Mengzhen Kou, Jinqiao Shi

Besides, our method allows researchers to distill knowledge from deeper networks to improve students further.

Knowledge Distillation

Document-Level Event Extraction via Human-Like Reading Process

no code implementations • 7 Feb 2022 • Shiyao Cui, Xin Cong, Bowen Yu, Tingwen Liu, Yucheng Wang, Jinqiao Shi

Meanwhile, rough reading is explored in a multi-round manner to discover undetected events, thus the multi-events problem is handled.

Document-level Event Extraction Event Extraction

Label Enhanced Event Detection with Heterogeneous Graph Attention Networks

no code implementations • 3 Dec 2020 • Shiyao Cui, Bowen Yu, Xin Cong, Tingwen Liu, Quangang Li, Jinqiao Shi

A heterogeneous graph attention networks is then introduced to propagate relational message and enrich information interaction.

Event Detection Graph Attention +1

Edge-Enhanced Graph Convolution Networks for Event Detection with Syntactic Relation

1 code implementation • Findings of the Association for Computational Linguistics 2020 • Shiyao Cui, Bowen Yu, Tingwen Liu, Zhen-Yu Zhang, Xuebin Wang, Jinqiao Shi

Previous studies on the task have verified the effectiveness of integrating syntactic dependency into graph convolutional networks.

Event Detection Relation