HFL at SemEval-2022 Task 8: A Linguistics-inspired Regression Model with Data Augmentation for Multilingual News Similarity

2022/04/11 by Zihang Xu, Ziqing Yang, Xu, Zihang +5 · 1 citation
Computer Science · #Natural Language Processing Techniques #Topic Modeling #Text and Document Classification Technologies

paper · pdf · doi:10.48550/arxiv.2204.04844

Abstract

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.

Cited by

• Hierarchical Level-Wise News Article Clustering via Multilingual Matryoshka Embeddings

Related

• BabelEnconding at SemEval-2020 Task 3: Contextual Similarity as a\n Combination of Multilingualism and Language Models
• Neobility at SemEval-2017 Task 1: An Attention-based Sentence Similarity Model
• A diverse Multilingual News Headlines Dataset from around the World
• fact check AI at SemEval-2025 Task 7: Multilingual and Crosslingual Fact-checked Claim Retrieval
• GateNLP-UShef at SemEval-2022 Task 8: Entity-Enriched Siamese Transformer for Multilingual News Article Similarity