Pairwise Supervised Contrastive Learning of Sentence Representations
This work improves sentence representation learning for natural language processing tasks, offering incremental advancements over prior methods.
The paper tackles the problem of sentence representation learning by addressing weaknesses in existing methods that fail to capture high-level semantic structure due to inadequate handling of contradiction pairs and local optima in loss functions, resulting in 10%–13% improvement on clustering tasks and 5%–6% improvement on semantic textual similarity tasks compared to previous state-of-the-art.
Many recent successes in sentence representation learning have been achieved by simply fine-tuning on the Natural Language Inference (NLI) datasets with triplet loss or siamese loss. Nevertheless, they share a common weakness: sentences in a contradiction pair are not necessarily from different semantic categories. Therefore, optimizing the semantic entailment and contradiction reasoning objective alone is inadequate to capture the high-level semantic structure. The drawback is compounded by the fact that the vanilla siamese or triplet losses only learn from individual sentence pairs or triplets, which often suffer from bad local optima. In this paper, we propose PairSupCon, an instance discrimination based approach aiming to bridge semantic entailment and contradiction understanding with high-level categorical concept encoding. We evaluate PairSupCon on various downstream tasks that involve understanding sentence semantics at different granularities. We outperform the previous state-of-the-art method with $10\%$--$13\%$ averaged improvement on eight clustering tasks, and $5\%$--$6\%$ averaged improvement on seven semantic textual similarity (STS) tasks.