More Discriminative Sentence Embeddings via Semantic Graph Smoothing
This work addresses the need for better sentence embeddings for document categorization tasks, but it is incremental as it builds on existing pretrained models with a smoothing technique.
The paper tackled the problem of learning more discriminative sentence representations in an unsupervised fashion by leveraging semantic graph smoothing to enhance embeddings from pretrained models, resulting in consistent improvements validated on eight benchmarks for text clustering and classification tasks.
This paper explores an empirical approach to learn more discriminantive sentence representations in an unsupervised fashion. Leveraging semantic graph smoothing, we enhance sentence embeddings obtained from pretrained models to improve results for the text clustering and classification tasks. Our method, validated on eight benchmarks, demonstrates consistent improvements, showcasing the potential of semantic graph smoothing in improving sentence embeddings for the supervised and unsupervised document categorization tasks.