Translates SemEval-2010 Task 8 to Romanian and evaluates Gemma 31B prompting and QLoRA fine-tuning against encoder baselines, finding fine-tuning reduces the cross-lingual gap to 1.4pp while smaller models perform within 1-4pp.
Relation classification via convolutional deep neural network,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Cross-lingual Relation Extraction with Large Language Models: Zero-Shot, Few-Shot, and Fine-Tuned Evaluation on Romanian
Translates SemEval-2010 Task 8 to Romanian and evaluates Gemma 31B prompting and QLoRA fine-tuning against encoder baselines, finding fine-tuning reduces the cross-lingual gap to 1.4pp while smaller models perform within 1-4pp.