Claim · 2014-09-01 · event · Stated in source · Supported

Bahdanau, Cho and Bengio (arXiv v1, 2014) propose letting a neural translation model (soft-)search for the parts of a source sentence relevant to each target word, instead of a fixed-length vector.

One recorded claim. The claim panel gives its qualification, the source passages with their place in each source, and how it was checked.

What it connects

Neural Machine Translation by Jointly Learning to Align and Translate → discusses →

Where it appears

No collection page holds this claim. It is listed with its records in the explorer.

Claim and evidence

Neural Machine Translation by Jointly Learning to Align and Translatediscusses ↓

Machine-checked, not reviewed by an editorStated in sourceSupported

Supporting evidence: primary · 1 source

Date: 2014-09-01 · event Records dated 2014 →

Scope and limitsThe v1 abstract does not use the word “attention” in the quoted passage; the link to the Transformer family rests on Vaswani and others’ citation of this paper, not on this text.

The evidence 01

primary sourceSupports the claim

Neural Machine Translation by Jointly Learning to Align and Translate

arXiv · 2014-09-01

Cite the original source

Cite the original document, not Selekoa. Check these details against it.

Source excerpt

“we conjecture that the use of a fixed-length vector is a bottleneck in improving the performance of this basic encoder–decoder architecture, and propose to extend this by allowing a model to automatically (soft-)search for parts of a source sentence that are relevant to predicting a target word”

Location in source PDF p. 1, abstract

How it was read Retrieved directly from the publisher

Machine-checked · 2026-10-31

Source details and check notes

Authors: Dzmitry Bahdanau; Yoshua Bengio
Version: arXiv:1409.0473v1, submitted 1 September 2014 (UTC). Retrieved for the candidate pack on 2026-10-28 (SHA-256 52391f14a0de04fdcbf2854e0349df5557ae01db0f37845e0aba96f0f1a9a2b1). The author list is the byline as printed on the first page.
Version date: 2014-09-01
Canonical address: Open the canonical source ↗
Accessed: 2026-10-28
Language: English (en)
Attribution: Dzmitry Bahdanau, Yoshua Bengio, “Neural Machine Translation by Jointly Learning to Align and Translate”, 2014-09-01.
Reuse: No open redistribution licence asserted. Bibliographic metadata, editorial paraphrases and short attributed quotations (at most 50 words each). Recheck terms before wider redistribution.
Check notes: Candidate pack passage, Kinds of AI and technical history (research/technical-history). The quotation (47 words) matched the text retrieved for the candidate pack on 2026-10-29. Matching ignored differences of whitespace, case, quotation marks, dashes, PDF line-break hyphens and Unicode compatibility forms, so the source’s typography can differ. No human editorial review.

Status and meaning

Editorial status. Not reviewed by an editor. An exploratory record that no editor has accepted. This status does not say whether a person has read any linked passage. Each evidence record above shows how it was checked.

Relation. The concept is explicitly discussed in the document. This makes no earliest-use, invention or transmission claim.

Date. Date of submission (an arXiv version or a journal’s received date, as the source states it); not a later revision or the venue’s publication date.

Record ID th-fam-tllm-bahdanau-2014