Summary of Relational Prompt-based Pre-trained Language Models For Social Event Detection, by Pu Li et al.
Relational Prompt-based Pre-trained Language Models for Social Event Detection
by Pu Li, Xiaoyan Yu, Hao Peng, Yantuan Xian, Linqin Wang, Li Sun, Jingyun Zhang, Philip S. Yu
First submitted to arxiv on: 12 Apr 2024
Categories
- Main: Computation and Language (cs.CL)
- Secondary: Artificial Intelligence (cs.AI); Machine Learning (cs.LG); Social and Information Networks (cs.SI)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary A novel approach to Social Event Detection (SED) is proposed, leveraging Pre-trained Language Models (PLMs) to address limitations of Graph Neural Network (GNN) based solutions. The RPLM_SED model employs a pairwise message modeling strategy to construct social messages into pairs with multi-relational sequences, and a prompt-based pairwise message learning mechanism to learn comprehensive message representations using PLMs. A clustering constraint is designed to optimize the encoding process, enhancing intra-cluster compactness and inter-cluster dispersion. Experimental results on three real-world datasets demonstrate state-of-the-art performance of RPLM_SED in offline, online, low-resource, and long-tail distribution scenarios for SED tasks. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Social Event Detection tries to find important events from what people are saying online. This helps with things like understanding public opinion and predicting risks. Some computers use a type of network called Graph Neural Networks (GNNs) to do this, but they can struggle when there’s missing or incorrect information between messages. Other methods don’t learn much from the words in the messages themselves. The new approach uses pre-trained language models to learn more about each message and how it relates to others. |
Keywords
* Artificial intelligence * Clustering * Event detection * Gnn * Graph neural network * Prompt
