HFL at SemEval-2022 Task 8: A Linguistics-inspired Regression Model with Data Augmentation for Multilingual News Similarity
This work addresses the challenge of accurately assessing similarity in multilingual news articles, which is important for applications like information retrieval and cross-lingual content analysis, but it appears incremental as it builds on existing techniques with task-specific adaptations.
The paper tackled the problem of measuring multilingual news article similarity by proposing a linguistics-inspired regression model with data augmentation, achieving first place on the SemEval-2022 Task 8 leaderboard with a Pearson's Correlation Coefficient of 0.818 on the official evaluation set.
This paper describes our system designed for SemEval-2022 Task 8: Multilingual News Article Similarity. We proposed a linguistics-inspired model trained with a few task-specific strategies. The main techniques of our system are: 1) data augmentation, 2) multi-label loss, 3) adapted R-Drop, 4) samples reconstruction with the head-tail combination. We also present a brief analysis of some negative methods like two-tower architecture. Our system ranked 1st on the leaderboard while achieving a Pearson's Correlation Coefficient of 0.818 on the official evaluation set.