Zuyu Zhao


2023

pdf bib
Unleashing the Power of Language Models in Text-Attributed Graph
Haoyu Kuang | Jiarong Xu | Haozhe Zhang | Zuyu Zhao | Qi Zhang | Xuanjing Huang | Zhongyu Wei
Findings of the Association for Computational Linguistics: EMNLP 2023

Representation learning on graph has been demonstrated to be a powerful tool for solving real-world problems. Text-attributed graph carries both semantic and structural information among different types of graphs. Existing works have paved the way for knowledge extraction of this type of data by leveraging language models or graph neural networks or combination of them. However, these works suffer from issues like underutilization of relationships between nodes or words or unaffordable memory cost. In this paper, we propose a Node Representation Update Pre-training Architecture based on Co-modeling Text and Graph (NRUP). In NRUP, we construct a hierarchical text-attributed graph that incorporates both original nodes and word nodes. Meanwhile, we apply four self-supervised tasks for different level of constructed graph. We further design the pre-training framework to update the features of nodes during training epochs. We conduct the experiment on the benchmark dataset ogbn-arxiv. Our method achieves outperformance compared to baselines, fully demonstrating its validity and generalization.

pdf bib
One-Model-Connects-All: A Unified Graph Pre-Training Model for Online Community Modeling
Ruoxue Ma | Jiarong Xu | Xinnong Zhang | Haozhe Zhang | Zuyu Zhao | Qi Zhang | Xuanjing Huang | Zhongyu Wei
Findings of the Association for Computational Linguistics: EMNLP 2023

Online community is composed of communities, users, and user-generated textual content, with rich information that can help us solve social problems. Previous research hasn’t fully utilized these three components and the relationship among them. What’s more, they can’t adapt to a wide range of downstream tasks. To solve these problems, we focus on a framework that simultaneously considers communities, users, and texts. And it can easily connect with a variety of downstream tasks related to social media. Specifically, we use a ternary heterogeneous graph to model online communities. Text reconstruction and edge generation are used to learn structural and semantic knowledge among communities, users, and texts. By leveraging this pre-trained model, we achieve promising results across multiple downstream tasks, such as violation detection, sentiment analysis, and community recommendation. Our exploration will improve online community modeling.

2022

pdf bib
Document-Level Relation Extraction via Pair-Aware and Entity-Enhanced Representation Learning
Xiusheng Huang | Hang Yang | Yubo Chen | Jun Zhao | Kang Liu | Weijian Sun | Zuyu Zhao
Proceedings of the 29th International Conference on Computational Linguistics

Document-level relation extraction aims to recognize relations among multiple entity pairs from a whole piece of article. Recent methods achieve considerable performance but still suffer from two challenges: a) the relational entity pairs are sparse, b) the representation of entity pairs is insufficient. In this paper, we propose Pair-Aware and Entity-Enhanced(PAEE) model to solve the aforementioned two challenges. For the first challenge, we design a Pair-Aware Representation module to predict potential relational entity pairs, which constrains the relation extraction to the predicted entity pairs subset rather than all pairs; For the second, we introduce a Entity-Enhanced Representation module to assemble directional entity pairs and obtain a holistic understanding of the entire document. Experimental results show that our approach can obtain state-of-the-art performance on four benchmark datasets DocRED, DWIE, CDR and GDA.