# Combining exogenous and endogenous signals with a semi-supervised co-attention network for early detection of COVID-19 fake tweets

arXiv: https://arxiv.org/abs/2104.05321 · HTML: https://mlsanity.com/paper/2104.05321/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- Builds on 15 of its 29 references that are in the map.
- First posted on arXiv 12 Apr 2021.

## Builds on (most cited first, 15 of 15)

- [RoBERTa: A Robustly Optimized BERT Pretraining Approach](https://mlsanity.com/paper/1907.11692/) (2019)
- [Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks](https://mlsanity.com/paper/1908.10084/) (2019)
- [Spectral Networks and Locally Connected Networks on Graphs](https://mlsanity.com/paper/1312.6203/) (2013)
- [Learning Convolutional Neural Networks for Graphs](https://mlsanity.com/paper/1605.05273/) (2016)
- [Adversarial Training Methods for Semi-Supervised Text Classification](https://mlsanity.com/paper/1605.07725/) (2016)
- [Hierarchical Question-Image Co-Attention for Visual Question Answering](https://mlsanity.com/paper/1606.00061/) (2016)
- [MixText: Linguistically-Informed Interpolation of Hidden Space for Semi-Supervised Text Classification](https://mlsanity.com/paper/2004.12239/) (2020)
- [Fake News Detection on Social Media using Geometric Deep Learning](https://mlsanity.com/paper/1902.06673/) (2019)
- [GCAN: Graph-aware Co-Attention Networks for Explainable Fake News Detection on Social Media](https://mlsanity.com/paper/2004.11648/) (2020)
- [Variational Pretraining for Semi-supervised Text Classification](https://mlsanity.com/paper/1906.02242/) (2019)
- [Weak Supervision for Fake News Detection via Reinforcement Learning](https://mlsanity.com/paper/1912.12520/) (2019)
- [Leveraging Multi-Source Weak Social Supervision for Early Detection of Fake News](https://mlsanity.com/paper/2004.01732/) (2020)
- [Cross-SEAN: A Cross-Stitch Semi-Supervised Neural Attention Model for COVID-19 Fake News Detection](https://mlsanity.com/paper/2102.08924/) (2021)
- [Semi-supervised Content-based Detection of Misinformation via Tensor Embeddings](https://mlsanity.com/paper/1804.09088/) (2018)
- [A Kernel of Truth: Determining Rumor Veracity on Twitter by Diffusion Pattern Alone](https://mlsanity.com/paper/2002.00850/) (2020)

## Built on by (most cited first, 5 of 5)

- [PCoT: Persuasion-Augmented Chain of Thought for Detecting Fake News and Social Media Disinformation](https://mlsanity.com/paper/2506.06842/) (2025)
- [MALicious INTent Dataset and Inoculating LLMs for Enhanced Disinformation Detection](https://mlsanity.com/paper/2603.14525/) (2026)
- [Checkovid: A COVID-19 misinformation detection system on Twitter using network and content mining perspectives](https://mlsanity.com/paper/2107.09768/) (2021)
- [Examining the Mental Health Impact of Misinformation on Social Media Using a Hybrid Transformer-Based Approach](https://mlsanity.com/paper/2503.02333/) (2025)
- [ExTax: Explainable Disinformation Detection via Persuasion, Emotion, and Narrative Role Taxonomies](https://mlsanity.com/paper/2605.27045/) (2026)

## Abstract (arXiv)

Fake tweets are observed to be ever-increasing, demanding immediate countermeasures to combat their spread. During COVID-19, tweets with misinformation should be flagged and neutralized in their early stages to mitigate the damages. Most of the existing methods for early detection of fake news assume to have enough propagation information for large labeled tweets -- which may not be an ideal setting for cases like COVID-19 where both aspects are largely absent. In this work, we present ENDEMIC, a novel early detection model which leverages exogenous and endogenous signals related to tweets, while learning on limited labeled data. We first develop a novel dataset, called CTF for early COVID-19 Twitter fake news, with additional behavioral test sets to validate early detection. We build a heterogeneous graph with follower-followee, user-tweet, and tweet-retweet connections and train a graph embedding model to aggregate propagation information. Graph embeddings and contextual features constitute endogenous, while time-relative web-scraped information constitutes exogenous signals. ENDEMIC is trained in a semi-supervised fashion, overcoming the challenge of limited labeled data. We propose a co-attention mechanism to fuse signal representations optimally. Experimental results on ECTF, PolitiFact, and GossipCop show that ENDEMIC is highly reliable in detecting early fake tweets, outperforming nine state-of-the-art methods significantly.