Contextualizing the Limits of Model & Evaluation Dataset Curation on Semantic Similarity Classification Tasks
It addresses the problem of unreliable performance evaluation in semantic similarity classification for practitioners deploying such systems in real-world contexts, though it is incremental as it reinforces prior work.
This paper examines how limitations in pre-trained models and evaluation datasets affect performance assessment in binary semantic similarity classification tasks, reinforcing prior findings of performance disparities across datasets, embedding techniques, and distance metrics.
This paper demonstrates how the limitations of pre-trained models and open evaluation datasets factor into assessing the performance of binary semantic similarity classification tasks. As (1) end-user-facing documentation around the curation of these datasets and pre-trained model training regimes is often not easily accessible and (2) given the lower friction and higher demand to quickly deploy such systems in real-world contexts, our study reinforces prior work showing performance disparities across datasets, embedding techniques and distance metrics, while highlighting the importance of understanding how data is collected, curated and analyzed in semantic similarity classification.