NLP Lineage

27 August 2019 · Representation and retrieval

Sentence-BERT

Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks

N. Reimers, I. Gurevych

Date
27 August 2019
What the date marks
arXiv submission
Kind
Research paper
Organisation
UKP Lab, TU Darmstadt
Access
Apache 2.0
Other dates
EMNLP 2019
Threads
In the guided tour
Chapter 10. Reading more than you remember
Learning phases
4. Embeddings and RAG

Notable: no later milestone in the data is documented as descending from it. How levels are decided

Read the primary source (arxiv.org)

Full summary in progress. Until then, this page shows the milestone's record from the research data.

What it changed

Before it

Comparing two sentences with BERT meant running the pair through the model together.

The problem

Finding the closest match in a corpus required a forward pass per candidate, which is hopeless at scale.

What it introduced

Train BERT in a siamese setup so each sentence maps to a fixed vector that can be compared with cosine similarity.

Why it matters

It turned BERT into an embedding model, which is what makes semantic search and RAG practical.

What modern models inherit

The embedding model as a separate component from the generator.

What came next

DPR for retrieval, and the sentence-transformers ecosystem most RAG stacks still use.

Before and after

The milestones it descends from, and the milestones that descend from it. Work developed alongside it is not counted. Weak links are left out unless you include them.

Before: where it came from

Sentence-BERT has 20 documented ancestors across 4 chapters, reaching back to Recurrent networks in 1990. Showing its direct parents and the 6 turning points among them.

Sentence-BERT has 20 documented ancestors across 4 chapters, reaching back to Recurrent networks in 1990. Showing all of them.

Sentence-BERT has 21 recorded ancestors, including indirect ones, across 4 chapters, reaching back to Recurrent networks in 1990. Showing its direct parents and the 6 turning points among them.

Sentence-BERT has 21 recorded ancestors, including indirect ones, across 4 chapters, reaching back to Recurrent networks in 1990. Showing all of them.

  1. 2019Sentence-BERTThis milestone.
  2. 2018BERT Turning pointSentence-BERT built directly on thisChapter 4. Pretrain once, use everywhere
  3. 2018GPT-1BERT was influenced by thisChapter 4. Pretrain once, use everywhere
  4. 2018ELMoBERT was influenced by thisChapter 4. Pretrain once, use everywhere
  5. 2018ULMFiTBERT was influenced by thisChapter 4. Pretrain once, use everywhere
  6. 2017CoVeELMo was influenced by thisChapter 3. Attention, then only attention
  7. 2017Transformer Turning pointBERT built directly on thisChapter 3. Attention, then only attention
  8. 2016Layer normTransformer built directly on thisChapter 3. Attention, then only attention
  9. 2015Residual connectionsTransformer was influenced by thisChapter 3. Attention, then only attention
  10. 2015LM pretraining Turning pointBERT was influenced by thisChapter 2. Words become vectors
  11. 2015Luong attentionTransformer was influenced by thisChapter 3. Attention, then only attention
  12. 2014AdamTransformer was influenced by thisChapter 2. Words become vectors
  13. 2014GloVeELMo built directly on thisChapter 2. Words become vectors
  14. 2014Seq2seq Turning pointTransformer built directly on thisChapter 2. Words become vectors
  15. 2014Attention Turning pointTransformer built directly on thisChapter 3. Attention, then only attention
  16. 2014GRU / encoder-decoderAttention built directly on thisChapter 2. Words become vectors
  17. 2013word2vecELMo was influenced by thisChapter 2. Words become vectors
  18. 2011NLP from scratchELMo was influenced by thisChapter 1. Language as probability
  19. 2003Neural LMNLP from scratch was influenced by this
    word2vec was influenced by this    Chapter 1. Language as probability
  20. 1997LSTM Turning pointELMo built directly on thisChapter 1. Language as probability
  21. 1990Recurrent networksLSTM was influenced by thisChapter 1. Language as probability
  1. 2019Sentence-BERTThis milestone.
  2. 2018BERT Turning pointSentence-BERT built directly on thisChapter 4. Pretrain once, use everywhere
  3. 2018GPT-1BERT was influenced by thisChapter 4. Pretrain once, use everywhere
  4. 2018ELMoBERT was influenced by thisChapter 4. Pretrain once, use everywhere
  5. 2018ULMFiTBERT was influenced by thisChapter 4. Pretrain once, use everywhere
  6. 2017CoVeELMo was influenced by thisChapter 3. Attention, then only attention
  7. 2017Transformer Turning pointBERT built directly on thisChapter 3. Attention, then only attention
  8. 2017ConvS2STransformer was influenced by this (indirect influence)Chapter 3. Attention, then only attention
  9. 2016Layer normTransformer built directly on thisChapter 3. Attention, then only attention
  10. 2015Residual connectionsTransformer was influenced by thisChapter 3. Attention, then only attention
  11. 2015LM pretraining Turning pointBERT was influenced by thisChapter 2. Words become vectors
  12. 2015Luong attentionTransformer was influenced by thisChapter 3. Attention, then only attention
  13. 2014AdamTransformer was influenced by thisChapter 2. Words become vectors
  14. 2014GloVeELMo built directly on thisChapter 2. Words become vectors
  15. 2014Seq2seq Turning pointTransformer built directly on thisChapter 2. Words become vectors
  16. 2014Attention Turning pointTransformer built directly on thisChapter 3. Attention, then only attention
  17. 2014GRU / encoder-decoderAttention built directly on thisChapter 2. Words become vectors
  18. 2013word2vecELMo was influenced by thisChapter 2. Words become vectors
  19. 2011NLP from scratchELMo was influenced by thisChapter 1. Language as probability
  20. 2003Neural LMNLP from scratch was influenced by this
    word2vec was influenced by this    Chapter 1. Language as probability
  21. 1997LSTM Turning pointELMo built directly on thisChapter 1. Language as probability
  22. 1990Recurrent networksLSTM was influenced by thisChapter 1. Language as probability

After: what it led to

No documented descendants are recorded for Sentence-BERT without weak links.

No documented descendants are recorded for Sentence-BERT without weak links.

No recorded descendants are recorded for Sentence-BERT.

No recorded descendants are recorded for Sentence-BERT.

Direct links and their sources

Where it came from

Built directly on

What it led to

Developed alongside