← All notes

Summary

BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding (2018) made bidirectional pretraining the default for natural language understanding. Instead of reading text only left-to-right, BERT is trained to fill in masked tokens using both sides of the context, then fine-tuned on a downstream task with a small task head.

What it introduced: masked language modeling and next-sentence prediction as pretraining objectives, plus a simple fine-tuning recipe that set new scores on the GLUE and SQuAD-style benchmarks of that period.

What it led to: a wave of encoder-only models, and later the sentence-embedding and retrieval work that still feeds RAG systems. See the pointer in this batch for Sentence-BERT, and live notes genai-rag-embeddings-vector-search and hf-transformers-auto-classes. The original Transformer pointer in this batch is the architecture BERT starts from.

Where to read it: https://arxiv.org/abs/1810.04805 (v2, last revised 24 May 2019).

The paper is posted on arXiv under the arXiv non-exclusive distribution license. This note links to it rather than reproducing it. The note itself is CC BY 4.0.

License: this note is CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Pointer by iinventedlight (AI-assisted). Paper license: arXiv non-exclusive distribution; this note links to it and does not reproduce it.

@misc{iil_note_xknpr931,
  author = {{'iinventedlight (AI-assisted)'}},
  title = {{BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding}},
  howpublished = {iinventedlight Library},
  year = {2026},
  date = {2026-11-11},
  url = {https://iinventedlight.com/library/notes/1810-04805-bert/},
  note = {Permanent ID iil:note:xknpr931. License: CC-BY-4.0.},
}

Markdown

Spotted a mistake? Use the MCP feedback tool or email hello@iinventedlight.com.

For language models and agents: start at /llms.txt | /agents.md | /.well-known/agent-library.json. This page in Markdown: /library/notes/1810-04805-bert.md.