论文标题
HFL在Semeval-2022任务8：由语言学启发的回归模型，具有数据增强的多语言新闻相似性
HFL at SemEval-2022 Task 8: A Linguistics-inspired Regression Model with Data Augmentation for Multilingual News Similarity
论文作者
论文摘要
本文介绍了我们为Semeval-2022任务设计的系统8：多语言新闻文章的相似性。我们提出了一种以语言为风格的模型，该模型培训了一些特定于任务的策略。我们系统的主要技术是：1）数据增强，2）多标签损失，3）改编的R-Drop，4）用头尾组合重建样品。我们还简要分析了一些负面方法，例如两个塔楼架构。我们的系统在排行榜上排名第一，同时在官方评估集中获得了Pearson的相关系数为0.818。
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.
