TL;DR
A transfer-cum-ensemble learning framework that integrates a task-specific pretrained model (XLM-RoBERTa) with a lightweight Indian-language model (IndicBERT) with a weighted-average attention mechanism is used to combine these sophisticated language models for further enhancing performance.
This study examines the effectiveness of two transformer-based architectures—BERT and DeBERTa—for identifying fake news using only textual information from headlines and article bodies and achieves strong performance on FakeDiverse corpus, demonstrating the need for enhanced generalization strategies as well as domain adaptation.
The proposed approach provides a simple and efficient solution for multilingual fake news detection in data-scarce environments with ensemble-based classifiers such as Random Forest and Gradient Boosting achieving reliable performance across both languages.
A hybrid transformer-based ensemble model for automated fake news identification using the FakeNewsNet dataset is proposed and Experimental results show that the ensemble model achieves an accuracy of approximately 93%, outperforming the individual constituent models.
In recent years, there has been an increase in the amount of fake news in the media, which is why fact-checking systems are gaining popularity, particularly those that use natural language processing (NLP) to quickly identify and flag fake news. One of the main limitations in the development of such systems is the limited number of datasets containing verified information, which are necessary for the effective training of models. The situation is particularly critical for non-English datasets, specifically those in the Ukrainian language. This article proposes a three-stage algorithm for training a model to recognize fake news in the Ukrainian language. At the core of the proposed approach lies the multilingual transformer model XLM-RoBERTa, which solves this problem by utilizing cross-lingual knowledge transfer from English to Ukrainian. This approach means there is no need to search for a large, high-quality dataset in Ukrainian; instead, a significantly smaller dataset in Ukrainian can be used for the final calibration of the model. The model developed as a result of the experiment proved effective in extreme low-resource scenarios, achieving 90.7% accuracy on just 500 training records and outperforming the baseline model by 9.7%.
The proposed framework highlights the potential of integrating transformer-based language models with classical machine learning algorithms to build robust and scalable fake news detection systems.
Disinformation on digital platforms is a very important problem for public trust and political debate. Even so, most research on automated fake news detection has stayed on a small number of high-resource languages. This paper presents AzFakeNews, the first large-scale benchmark dataset for fake news detection in Azerbaijani. Azerbaijani is a low-resource Turkic language with more than 30 million speakers. We built the dataset from two sources: scraping of news articles from five major Azerbaijani media websites, and generation of synthetic fake articles via Meta's LLaMA language model. The final dataset has 3,782 articles (3,000 authentic and 782 synthetic) across 34 topic categories. For the baseline we fine-tuned the Azerbaijani aLLMA model on this corpus. It reached 91.03% accuracy and 91.11% macro-F1 on the test split, because of which we consider it a strong baseline compared to mBERT, XLM-RoBERTa and the base LLaMA on the same data. We hope the dataset helps close a gap in Azerbaijani NLP and supports cross-language research on disinformation.
We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.
