AI ChronicleAI 纪年 Full timeline →

Research breakthrough · curated entry

Sequence to sequence learning makes end-to-end translation possible

Sequence to Sequence Learning with Neural Networks

Date
September 10, 2014
Category
Research breakthrough
Organization / Model
Google
Evidence
1 source · 1 evidence family

Sutskever and colleagues proposed mapping a variable-length input to a variable-length output with two recurrent networks. It became the key architectural basis for machine translation and later for dialogue systems.

Curator's note: Cross-checked against the arXiv paper.

Curated sources · 1

  1. Paper preprints · arXiv2014-09-10

    Sequence to Sequence Learning with Neural Networks

    arxiv.org

Cite this page

This page is a curated static snapshot and can be cited directly. Preferred format GB/T 7714—2025 (electronic resource [EB/OL]); the other 5 styles open below, together with the machine-readable files.

GB/T 7714—2025 (electronic resource [EB/OL])

AI 纪年. 序列到序列学习提出,端到端翻译成为可能[EB/OL]. (2014-09-10)[2026-11-03]. https://aihist.net/e/seq2seq-2014.html.

APA 7 · American Psychological Association

AI 纪年. (2014, September 10). 序列到序列学习提出,端到端翻译成为可能 [Web page]. https://aihist.net/e/seq2seq-2014.html

MLA 9 · Modern Language Association

"序列到序列学习提出,端到端翻译成为可能." AI 纪年, 10 Sept. 2014, aihist.net/e/seq2seq-2014.html. Accessed 3 Nov. 2026.

Chicago 17 · notes-bibliography

AI 纪年. "序列到序列学习提出,端到端翻译成为可能." Published Sept. 10, 2014. Accessed Nov. 3, 2026. https://aihist.net/e/seq2seq-2014.html.

IEEE · citation-reference list

"序列到序列学习提出,端到端翻译成为可能," AI 纪年, 10 Sept. 2014. [Online]. Available: https://aihist.net/e/seq2seq-2014.html. [Accessed: 3 Nov. 2026].

ACM · author-date variant

AI 纪年. September 10, 2014. 序列到序列学习提出,端到端翻译成为可能. Retrieved November 3, 2026 from https://aihist.net/e/seq2seq-2014.html.

All six styles come from one template and can be compared word for word. [引用日期], Accessed, urldate and Y2 all carry this page's build date, 2026-11-03 — a static page is an immutable snapshot, so substitute the date you actually consulted it. The address in every style is the archive's canonical page: this English page translates the same snapshot, and the work keeps its name, so one record cites both. For one-click copying use the same entry in the timeline, which takes the day you opened it. Expand the other styles before printing.