Transformers

Atefeh Hemmati, Amir Masoud Rahmani, Fatemeh Bazikar, Hossein Moosaei and Panos M. Pardalos
Additional contact information
Atefeh Hemmati: Islamic Azad University, Department of Computer Engineering, SR.C
Amir Masoud Rahmani: National Yunlin University of Science and Technology, Future Technology Research Center
Fatemeh Bazikar: Alzahra University, Department of Computer Science, Faculty of Mathematical Sciences
Hossein Moosaei: Jan Evangelista Purkyně University, Department of Informatics, Faculty of Science
Panos M. Pardalos: University of Florida, Department of Industrial & Systems Engineering

Chapter Chapter 12 in Optimization Techniques for Deep Learning, 2026, pp 165-178 from Springer

Abstract: Abstract On 2017, eight researchers dropped a bomb. Not with guns. With eight words: Attention is all you need. And just like that, RNNs became obsolete. LSTMs became legacy. The future of AI had a new engine. Transformers did not evolve sequence modeling. They reinvented it. No loops. No recurrence. Just pure, parallel attention, every token talking to every other, all at once. This is the architecture behind ChatGPT, BERT, DALL·E, AlphaFold, the backbone of modern AI.

Date: 2026
References: Add references at CitEc
Citations:

There are no downloads for this item, see the EconPapers FAQ for hints about obtaining it.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:spr:spochp:978-3-032-20703-6_12

Ordering information: This item can be ordered from
http://www.springer.com/9783032207036

DOI: 10.1007/978-3-032-20703-6_12

Access Statistics for this chapter

More chapters in Springer Optimization and Its Applications from Springer
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().