LLMs in Combating Disinformation: The Influence of Model Size on Fake News Detection in Brazil
Abstract
Disinformation and the spread of fake news have significant negative impacts across various domains, particularly on democracy, as they can be used to manipulate public opinion during electoral processes. Although the scientific literature presents studies aimed at detecting fake news, there are still considerable gaps regarding the Portuguese (Brazilian) language. In this context, this study investigates the hypothesis that the number of parameters in LLMs (Large Language Models) influences the detection of fake news. The results suggest that the hypothesis is valid. Additionally, other influencing factors were observed, as well as a performance limit inherent to the analyzed characteristics.
References
Gôlo, M. P. S., Mori, A. L. V., Oliveira, W. G., Barbosa, J. R., GracianoNeto, V. V., de Lima, E. A., and Marcacini, R. M. (2023). On the use of Large Language Models to Detect Brazilian Politics Fake News. In Proceedings of the Brazilian Symposium on Artificial Intelligence (SBIA) 2023, pages 1–12. Sociedade Brasileira de Computação (SBC).
Hu, B., Sheng, Q., Cao, J., Shi, Y., Li, Y., Wang, D., and Qi, P. (2024). Bad actor, good advisor: Exploring the role of large language models in fake news detection. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 38.
Pelrine, A. e. a. (2023). Towards reliable misinformation mitigation: Generalization, uncertainty, and gpt-4. Arxiv.
Qu, Z., Meng, Y., Muhammad, G., and Tiwari, P. (2024). Qmfnd: A quantum multimodal fusion-based fake news detection model for social media. Information Fusion.
Roumeliotis, K. I., Tselikas, N. D., and Nasiopoulos, D. K. (2025). Fake news detection and classification: A comparative study of convolutional neural networks, large language models, and natural language processing models. Future Internet, 17(1).
Hu, B., Sheng, Q., Cao, J., Shi, Y., Li, Y., Wang, D., and Qi, P. (2024). Bad actor, good advisor: Exploring the role of large language models in fake news detection. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 38.
Pelrine, A. e. a. (2023). Towards reliable misinformation mitigation: Generalization, uncertainty, and gpt-4. Arxiv.
Qu, Z., Meng, Y., Muhammad, G., and Tiwari, P. (2024). Qmfnd: A quantum multimodal fusion-based fake news detection model for social media. Information Fusion.
Roumeliotis, K. I., Tselikas, N. D., and Nasiopoulos, D. K. (2025). Fake news detection and classification: A comparative study of convolutional neural networks, large language models, and natural language processing models. Future Internet, 17(1).
Published
2025-09-29
How to Cite
SCHETTINI, Pedro T.; POSSARI, Guilherme; FAIÇAL, Bruno S.; BARROS, Rodolfo M..
LLMs in Combating Disinformation: The Influence of Model Size on Fake News Detection in Brazil. In: BRAZILIAN SYMPOSIUM IN INFORMATION AND HUMAN LANGUAGE TECHNOLOGY (STIL), 16. , 2025, Fortaleza/CE.
Anais [...].
Porto Alegre: Sociedade Brasileira de Computação,
2025
.
p. 723-727.
DOI: https://doi.org/10.5753/stil.2025.37879.
