We are a research team from the Advanced Knowledge Center in Immersive Technologies (AKCIT) at the Federal University of Goiás (UFG). Our work is dedicated to combating misinformation by developing advanced computational techniques for fact-checking.
This initiative is part of the broader project, "Computational Techniques for Multimodal Data Security and Privacy", running from July 2024 to December 2025.
Our team includes:
- 4 AI Bachelor's Students from UFG: @Fabrycio-Nakano, @kauandivino,@Maykon e @victoremanuelgo.
- 1 Computer Science Master's Student: @ju-resplande
- Project Lead: Prof. Arlindo Galvão
AKCIT-FN at CheckThat! 2025: Switching Fine-Tuned SLMs and LLM Prompting for Multilingual Claim Normalization
- In this paper, we detail our hybrid model that achieved top-three positions in 15 of the 20 languages in the competition.
- Our LLM-based zero-shot strategy proved highly effective, securing second place in five of the seven languages that had no prior training data.
- Conference: CheckThat! Lab at CLEF 2025
- Links: 📄 Read the Paper | 💻 View Code
- This work addresses the resource gap in Portuguese Automated Fact-Checking (AFC).
- We introduce a novel pipeline to systematically enrich misinformation datasets with external web evidence 🔎, enhancing their reliability and verifiability.
- Conference: 1st Workshop on Fact-Checking and Trustworthiness (FEVER 2025)
- Links: 📄 Read the Paper | 💻 View Code
This work has been fully funded by the project "Computational Techniques for Multimodal Data Security and Privacy" supported by the Advanced Knowledge Center in Immersive Technologies (AKCIT), with financial resources from the PPI IoT/Manufatura 4.0 / PPI HardwareBR of the MCTI grant number 057/2023, signed with EMBRAPII.