CrossNews-UA: A Cross-lingual News Semantic Similarity Benchmark for Ukrainian, Polish, Russian, and English
Daryna Dementieva, Evgeniya Sukhodolskaya, Alexander Fraser
TL;DR
The paper tackles the challenge of multilingual news verification beyond English by introducing CrossNews-UA, a scalable crowdsourced benchmark for cross-lingual news semantic similarity across Ukrainian, Polish, Russian, and English. It presents a 4W-based explainable annotation scheme, a comprehensive data-collection pipeline (including automated Ukrainian scraping and crowdsourced labeling with LLM-assisted quality control), and a final dataset of 500 news pairs. The authors evaluate a range of baselines—from traditional embeddings (Bag-of-Words, BERT, XLM-RoBERTa, mT5, E5-large) to prompting-based LLMs (Mistral, Llama3, Aya-23)—revealing key insights such as the relative strength of certain encoders, the surprising efficacy of simple BoW in some settings, and the persistent difficulty of temporal/When alignment. They also discuss limitations like fixed 2022-topic coverage and dataset size, and offer a pathway for broader, more explainable cross-lingual news analysis with underrepresented languages in mind.
Abstract
In the era of social networks and rapid misinformation spread, news analysis remains a critical task. Detecting fake news across multiple languages, particularly beyond English, poses significant challenges. Cross-lingual news comparison offers a promising approach to verify information by leveraging external sources in different languages (Chen and Shu, 2024). However, existing datasets for cross-lingual news analysis (Chen et al., 2022a) were manually curated by journalists and experts, limiting their scalability and adaptability to new languages. In this work, we address this gap by introducing a scalable, explainable crowdsourcing pipeline for cross-lingual news similarity assessment. Using this pipeline, we collected a novel dataset CrossNews-UA of news pairs in Ukrainian as a central language with linguistically and contextually relevant languages-Polish, Russian, and English. Each news pair is annotated for semantic similarity with detailed justifications based on the 4W criteria (Who, What, Where, When). We further tested a range of models, from traditional bag-of-words, Transformer-based architectures to large language models (LLMs). Our results highlight the challenges in multilingual news analysis and offer insights into models performance.