Comparer des méthodes
Examinez les méthodes sélectionnées côte à côte ; les lignes qui diffèrent sont mises en évidence.
| Embeddings de phrases faiblement supervisées× | Plongements de phrases auto-supervisés× | |
|---|---|---|
| Domaine | Apprentissage profond | Apprentissage profond |
| Famille | Machine learning | Machine learning |
| Année d'origine≠ | 2016–2019 | 2019–2021 |
| Auteur d'origine≠ | Ratner et al. (weak supervision framework); Reimers & Gurevych (sentence embeddings) | Gao, T., Yao, X., & Chen, D. (SimCSE); Reimers, N. & Gurevych, I. (Sentence-BERT) |
| Type≠ | Representation learning under weak supervision | Self-supervised representation learning |
| Source fondatrice≠ | Ratner, A., De Sa, C., Wu, S., Selsam, D., & Re, C. (2016). Data Programming: Creating Large Training Sets, Quickly. Advances in Neural Information Processing Systems (NeurIPS), 29. link ↗ | Gao, T., Yao, X., & Chen, D. (2021). SimCSE: Simple Contrastive Learning of Sentence Embeddings. Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing (EMNLP), 6894–6910. DOI ↗ |
| Alias | WS sentence embeddings, noisy-label sentence representation learning, weakly supervised sentence representation, distant-supervision sentence embeddings | self-supervised sentence representation learning, contrastive sentence embeddings, SimCSE, unsupervised sentence encoders |
| Apparentées≠ | 6 | 5 |
| Résumé≠ | Weakly supervised sentence embeddings train dense sentence representations using noisy, heuristic, or programmatically generated labels instead of costly human annotation. Labeling functions — rules, distant supervision signals, or lightweight classifiers — supply approximate supervision that a label model aggregates into probabilistic labels, which then guide the sentence encoder to produce task-useful representations at scale. | Self-supervised sentence embeddings train a neural encoder to map sentences into a dense vector space without requiring manually labeled pairs. By constructing positive examples automatically — for instance by passing the same sentence through dropout twice — and using contrastive objectives, the model learns semantically rich representations that transfer well to similarity, retrieval, and classification tasks. |
| ScholarGateJeu de données ↗ |
|
|