Scaling Projection-Based Cross-Lingual NER with Image-Text Pairs in XTREME-R Low-Resource Languages
Domain:
natural language processingCreator:
Ass
Publisher:
Host:
Cross-lingual Named Entity Recognition (NER) leverages knowledge transfer between languages to identify and classify named entities, making it particularly useful for low-resource languages. We show that the data-based cross-lingual transfer method is an effective technique for crosslingual NER and can outperform multilingual language models for low-resource languages. This paper introduces two key enhancements to the annotation projection step in cross-lingual NER for low-resource languages. First, we explore refining word alignments using back-translation to improve accuracy. Second, we pres
Research goal: How does the performance of projection-based cross-lingual NER with image-text pairs scale with varying amounts of available parallel data in XTREME-R low-resource languages, compared to text-only baselines?
Autonomous synthesis report generated by Assignee Research. Tribunal consensus score: 7.5/10. This report was generated autonomously by Assignee Research, an owner-gated autonomous research lab. The content synthesizes findings from peer-reviewed papers. Tribunal score: 7.5/10.
Similar
Cross-lingual NER performance variation in XTREME-R low-resource languages via annotation projectionProjection-based Cross-lingual NER Scaling and F1-Score Gaps in Low-Resource LanguagesPerformance comparison of projection-based cross-lingual NER and few-shot multilingual LLMs on XTREME-R low-resource languagesImpact of Annotation Projection Quality on Cross-Lingual NER Accuracy in XTREME-R Low-Resource Languages versus mT5 and XLM-RProjection-based Method Scalability and Cross-lingual NER Accuracy in Low-Resource LanguagesMitigating Annotation Projection Noise in Cross-Lingual NER via Source Dataset Scaling for Low-Resource Languages