MultiMind at SemEval-2025 Task 7: Crosslingual Fact-Checked Claim Retrieval via Multi-Source Alignment

TL;DR

This paper introduces TriAligner, a dual-encoder system utilizing contrastive learning and multi-source translation to improve crosslingual claim retrieval for fact-checking, showing significant accuracy gains.

Contribution

We propose TriAligner, a novel multi-source alignment method that effectively retrieves claims across languages using dual-encoder architecture and contrastive learning.

Findings

Significant improvement in retrieval accuracy over baselines.

Effective crosslingual claim retrieval demonstrated on benchmarks.

Robustness enhanced through data augmentation and hard negative sampling.

Abstract

This paper presents our system for SemEval-2025 Task 7: Multilingual and Crosslingual Fact-Checked Claim Retrieval. In an era where misinformation spreads rapidly, effective fact-checking is increasingly critical. We introduce TriAligner, a novel approach that leverages a dual-encoder architecture with contrastive learning and incorporates both native and English translations across different modalities. Our method effectively retrieves claims across multiple languages by learning the relative importance of different sources in alignment. To enhance robustness, we employ efficient data preprocessing and augmentation using large language models while incorporating hard negative sampling to improve representation learning. We evaluate our approach on monolingual and crosslingual benchmarks, demonstrating significant improvements in retrieval accuracy and fact-checking performance over baselines.