fact check AI at SemEval-2025 Task 7: Multilingual and Crosslingual Fact-checked Claim Retrieval
This addresses the problem of retrieving fact-checked claims across languages for users in misinformation detection, but it is incremental as it applies existing methods to a new competition task.
The paper tackled multilingual and crosslingual fact-checked claim retrieval by fine-tuning a bi-encoder transformer model for Learning-to-Rank, achieving 92% Success@10 in multilingual and 80% Success@10 in crosslingual retrieval.
SemEval-2025 Task 7: Multilingual and Crosslingual Fact-Checked Claim Retrieval is approached as a Learning-to-Rank task using a bi-encoder model fine-tuned from a pre-trained transformer optimized for sentence similarity. Training used both the source languages and their English translations for multilingual retrieval and only English translations for cross-lingual retrieval. Using lightweight models with fewer than 500M parameters and training on Kaggle T4 GPUs, the method achieved 92% Success@10 in multilingual and 80% Success@10 in 5th in crosslingual and 10th in multilingual tracks.