Cross-lingual Relation Extraction with Large Language Models: Zero-Shot, Few-Shot, and Fine-Tuned Evaluation on Romanian

📰 ArXiv cs.AI

Learn how to apply large language models to cross-lingual relation extraction for low-resource languages like Romanian, and evaluate their performance in zero-shot, few-shot, and fine-tuned settings.

advanced Published 1 Jul 2026
Action Steps
  1. Translate an English benchmark dataset to a low-resource language like Romanian using an LLM-based translation pipeline.
  2. Evaluate a large language model like Gemma 4 31B on the translated dataset in zero-shot, few-shot, and fine-tuned settings.
  3. Apply QLoRA fine-tuning to improve the model's performance on the target language.
  4. Compare the results of different evaluation settings to determine the most effective approach for cross-lingual relation extraction.
  5. Use the insights gained from this study to develop more accurate relation extraction models for low-resource languages.
Who Needs to Know This

NLP engineers and researchers working on low-resource languages can benefit from this study to improve relation extraction performance. The findings can be applied to develop more accurate models for languages with limited annotated corpora.

Key Insight

💡 Large language models can be effectively used for cross-lingual relation extraction in low-resource languages, with fine-tuning and few-shot learning leading to significant performance improvements.

Share This
✅ Improve relation extraction for low-resource languages like Romanian with large language models! ✅

Key Takeaways

Learn how to apply large language models to cross-lingual relation extraction for low-resource languages like Romanian, and evaluate their performance in zero-shot, few-shot, and fine-tuned settings.

Full Article

Title: Cross-lingual Relation Extraction with Large Language Models: Zero-Shot, Few-Shot, and Fine-Tuned Evaluation on Romanian

Abstract:
arXiv:2606.31718v1 Announce Type: cross Abstract: Relation extraction (RE) for low-resource languages is typically constrained by the lack of annotated corpora. We investigate the feasibility of cross-lingual RE for Romanian by combining automatic dataset translation with large language model (LLM) inference. We translate the SemEval-2010 Task 8 benchmark from English to Romanian using an LLM-based translation pipeline and evaluate Gemma 4 31B under zero-shot, few-shot, and QLoRA fine-tuned conf
Read full paper → ← Back to Reads

Related Videos

5 Levels of AI Agents - From Simple LLM Calls to Multi-Agent Systems
5 Levels of AI Agents - From Simple LLM Calls to Multi-Agent Systems
Dave Ebbelaar (LLM Eng)
I Tested My AI-Powered Autocoder With 3 Different LLM Models
I Tested My AI-Powered Autocoder With 3 Different LLM Models
Making Made Easy
You Can Run Your Own Powerful LLM AI On Almost Any Computer! OPEN SOURCE! NO GPU NEEDED! MISTRAL 7B!
You Can Run Your Own Powerful LLM AI On Almost Any Computer! OPEN SOURCE! NO GPU NEEDED! MISTRAL 7B!
Making Made Easy
How To Run Mistral 7B LLM AI At Full Precision On A Raspberry Pi 5 With 4GB Of RAM #Overload
How To Run Mistral 7B LLM AI At Full Precision On A Raspberry Pi 5 With 4GB Of RAM #Overload
Making Made Easy
Google's Secret AI That's 10X More Powerful Than ChatGPT
Google's Secret AI That's 10X More Powerful Than ChatGPT
Kevin Farugia AI Automation
Notebook LM New Video Capabilities - Is It Overrated?
Notebook LM New Video Capabilities - Is It Overrated?
Kevin Farugia AI Automation