Manual Evaluation Matters: Reviewing Test Protocols of Distantly Supervised Relation Extraction
This addresses a critical evaluation issue for researchers in distantly supervised relation extraction, providing more credible benchmarks and insights to advance the field.
The paper tackled the problem of inaccurate evaluation in distantly supervised relation extraction by building manually-annotated test sets for NYT10 and Wiki20 datasets, revealing that automatic evaluation can produce up to 53% wrong labels and lead to different conclusions, such as pre-trained models achieving dominating performance but being more susceptible to false-positives.
Distantly supervised (DS) relation extraction (RE) has attracted much attention in the past few years as it can utilize large-scale auto-labeled data. However, its evaluation has long been a problem: previous works either took costly and inconsistent methods to manually examine a small sample of model predictions, or directly test models on auto-labeled data -- which, by our check, produce as much as 53% wrong labels at the entity pair level in the popular NYT10 dataset. This problem has not only led to inaccurate evaluation, but also made it hard to understand where we are and what's left to improve in the research of DS-RE. To evaluate DS-RE models in a more credible way, we build manually-annotated test sets for two DS-RE datasets, NYT10 and Wiki20, and thoroughly evaluate several competitive models, especially the latest pre-trained ones. The experimental results show that the manual evaluation can indicate very different conclusions from automatic ones, especially some unexpected observations, e.g., pre-trained models can achieve dominating performance while being more susceptible to false-positives compared to previous methods. We hope that both our manual test sets and novel observations can help advance future DS-RE research.