USTCCTSU at SemEval-2024 Task 1: Reducing Anisotropy for Cross-lingual Semantic Textual Relatedness Task

TL;DR

This paper presents a method for cross-lingual semantic relatedness that reduces anisotropy in sentence representations using whitening and data filtering, achieving top results in SemEval-2024 Task 1.

Contribution

It introduces a whitening-based approach and a data filtering technique to improve cross-lingual sentence embeddings, leading to enhanced performance in multilingual relatedness tasks.

Findings

Achieved 2nd place in Spanish relatedness

Secured 3rd place in Indonesian relatedness

Multiple top-ten results in the competition's track C

Abstract

Cross-lingual semantic textual relatedness task is an important research task that addresses challenges in cross-lingual communication and text understanding. It helps establish semantic connections between different languages, crucial for downstream tasks like machine translation, multilingual information retrieval, and cross-lingual text understanding.Based on extensive comparative experiments, we choose the XLM-R-base as our base model and use pre-trained sentence representations based on whitening to reduce anisotropy.Additionally, for the given training data, we design a delicate data filtering method to alleviate the curse of multilingualism. With our approach, we achieve a 2nd score in Spanish, a 3rd in Indonesian, and multiple entries in the top ten results in the competition's track C. We further do a comprehensive analysis to inspire future research aimed at improving performance on cross-lingual tasks.