方法对比

	可解释句子嵌入	自监督句子嵌入
领域	深度学习	深度学习
方法族	Machine learning	Machine learning
起源年份	2016–2018	2019–2021
提出者	Conneau et al.; Ribeiro et al. (probing + LIME frameworks)	Gao, T., Yao, X., & Chen, D. (SimCSE); Reimers, N. & Gurevych, I. (Sentence-BERT)
类型	Post-hoc interpretability applied to sentence encoders	Self-supervised representation learning
开创性文献	Conneau, A., Kruszewski, G., Lample, G., Barrault, L., & Baroni, M. (2018). What you can cram into a single $\vec{v}$ector: Probing sentence embeddings for linguistic properties. In Proceedings of ACL 2018, pp. 2126–2136.	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.
别名	interpretable sentence representations, XAI sentence embeddings, probing sentence embeddings, explainable sentence vectors	self-supervised sentence representation learning, contrastive sentence embeddings, SimCSE, unsupervised sentence encoders
相关	6	5
摘要	Explainable sentence embeddings combine dense sentence representation learning with post-hoc or intrinsic interpretability tools — such as probing classifiers, LIME, SHAP, or attention attribution — to reveal what linguistic and semantic information is encoded in a sentence vector and why a downstream model makes a given prediction. The goal is to retain the representational power of modern encoders while making their behavior auditable.	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.