Team Trifecta at Factify5WQA: Setting the Standard in Fact Verification with Fine-Tuning
This work addresses fact verification for misinformation detection, representing an incremental advance by combining existing methods like fine-tuning and ensembles in a competition setting.
The paper tackled fact verification by developing Pre-CoFactv3, a framework combining question answering and text classification with fine-tuned LLMs and FakeNet, achieving first place in the AAAI-24 Factify 3.0 Workshop with a 103% accuracy improvement over the baseline and a 70% lead over the second competitor.
In this paper, we present Pre-CoFactv3, a comprehensive framework comprised of Question Answering and Text Classification components for fact verification. Leveraging In-Context Learning, Fine-tuned Large Language Models (LLMs), and the FakeNet model, we address the challenges of fact verification. Our experiments explore diverse approaches, comparing different Pre-trained LLMs, introducing FakeNet, and implementing various ensemble methods. Notably, our team, Trifecta, secured first place in the AAAI-24 Factify 3.0 Workshop, surpassing the baseline accuracy by 103% and maintaining a 70% lead over the second competitor. This success underscores the efficacy of our approach and its potential contributions to advancing fact verification research.