A Structured Self-Attentive Sentence Embedding
Release time:2026-08-26
Hits:
- Journal:
- ICLR, 2017
- Abstract:
- Zhouhan Lin, Minwei Feng, Cicero Nogueira dos Santos, Mo Yu, Bing Xiang, Bowen Zhou†, Yoshua Bengio†.
- Note:
- Pioneering First task-agnostic self-attention and multi-head mechanisms — adopted by the Transformer, the foundation of GPT, BERT, and all modern LLMs.
- Translation or Not:
- no
- Links to published journals:
- https://arxiv.org/abs/1703.03130