Sentence embedding: Difference between revisions
RobowaifuDev (talk | contribs) m (Added references heading) |
RobowaifuDev (talk | contribs) m (Added DiffCSE) |
||
| Line 3: | Line 3: | ||
=== State of the art === | === State of the art === | ||
As of October 2022, [[CLIP]] text embeddings from a 38M parameter model have been found out perform [[BERT]] and Phrase-BERT 110M parameter models, when using [[domain aware prompting]] on sentences from news articles ([[CoNLL-2003]]), chemical-disease interactions ([[BC5CDR]]), and emerging and rare entity recognition ([[WNUT 2017]]).<ref>An Yan, Jiacheng Li, Wanrong Zhu, Yujie Lu, William Yang Wang, Julian McAuley. "CLIP also Understands Text: Prompting CLIP for Phrase Understanding." 2022; [https://arxiv.org/abs/2210.05836 arXiv:2210.05836]</ref> Without domain aware prompting, CLIP still outperformed other models on sentences from news articles. | As of October 2022, [[CLIP]] text embeddings from a 38M parameter model have been found out perform [[BERT]] and Phrase-BERT 110M parameter models, when using [[domain aware prompting]] on sentences from news articles ([[CoNLL-2003]]), chemical-disease interactions ([[BC5CDR]]), and emerging and rare entity recognition ([[WNUT 2017]]).<ref>An Yan, Jiacheng Li, Wanrong Zhu, Yujie Lu, William Yang Wang, Julian McAuley. "CLIP also Understands Text: Prompting CLIP for Phrase Understanding." 2022; [https://arxiv.org/abs/2210.05836 arXiv:2210.05836]</ref> Without domain aware prompting, CLIP still outperformed other models on sentences from news articles. | ||
As of April 2022, [[DiffCSE]] achieves state-of-the-art results in unsupervised sentence representation learning.<ref>Chuang et al. "DiffCSE: Difference-based Contrastive Learning for Sentence Embeddings." 2022. [https://arxiv.org/abs/2204.10298 arXiv:2204.10298]</ref> | |||
=== Pretrained models === | === Pretrained models === | ||
Latest revision as of 03:21, 9 January 2023
A sentence embedding is a technique in natural language processing where sentences are mapped to vectors and can be used for similarity search. In transformer models this is usually achieved with a classification token but it can also be done by taking the first token of the hidden state of a transformer encoder or mean pooling over all tokens, from the last layer or multiple layers.
State of the art
As of October 2022, CLIP text embeddings from a 38M parameter model have been found out perform BERT and Phrase-BERT 110M parameter models, when using domain aware prompting on sentences from news articles (CoNLL-2003), chemical-disease interactions (BC5CDR), and emerging and rare entity recognition (WNUT 2017).[1] Without domain aware prompting, CLIP still outperformed other models on sentences from news articles.
As of April 2022, DiffCSE achieves state-of-the-art results in unsupervised sentence representation learning.[2]
Pretrained models
Sentence transformers provides a variety of pretrained models for sentence embeddings.
References
- ↑ An Yan, Jiacheng Li, Wanrong Zhu, Yujie Lu, William Yang Wang, Julian McAuley. "CLIP also Understands Text: Prompting CLIP for Phrase Understanding." 2022; arXiv:2210.05836
- ↑ Chuang et al. "DiffCSE: Difference-based Contrastive Learning for Sentence Embeddings." 2022. arXiv:2204.10298