Varvara Arzt
@kleines-gespenst.bsky.social
Doing PhD in LM interpretability at TU Wien. Views are mine ☮️
TL;DR: We examine the OOD performance of RE models focusing on biographical relation overlap across general-purpose RE datasets and find that data quality dictates adaptation strategy: fine-tuning excels with clean data, ICL with noisy data. Zero-shot sometimes outperforms all! [2/5]
May 20, 2025 at 1:47 PM
TL;DR: We examine the OOD performance of RE models focusing on biographical relation overlap across general-purpose RE datasets and find that data quality dictates adaptation strategy: fine-tuning excels with clean data, ICL with noisy data. Zero-shot sometimes outperforms all! [2/5]
Relation Extraction or Pattern Matching? How well do RE models generalise to OOD data? We find that higher in-distribution scores do not necessarily translate to better transferability.
Paper: arxiv.org/abs/2505.12533
Paper: arxiv.org/abs/2505.12533
May 20, 2025 at 1:47 PM
Relation Extraction or Pattern Matching? How well do RE models generalise to OOD data? We find that higher in-distribution scores do not necessarily translate to better transferability.
Paper: arxiv.org/abs/2505.12533
Paper: arxiv.org/abs/2505.12533