Papers › The Evolved Transformer

The Evolved Transformer

30 Jan 2019arXiv:1901.11117archive 2025-07-28

David R. So, Chen Liang, Quoc V. Le

Recent works have highlighted the strength of the Transformer architecture on sequence tasks while, at the same time, neural architecture search (NAS) has begun to outperform human-designed models. Our goal is to apply NAS to search for a better alternative to the Transformer. We first construct a large search space inspired by the recent advances in feed-forward sequence models and then run evolutionary architecture search with warm starting by seeding our initial population with the Transformer. To directly search on the computationally expensive WMT 2014 English-German translation task, we develop the Progressive Dynamic Hurdles method, which allows us to dynamically allocate more resources to more promising candidate models. The architecture found in our experiments -- the Evolved Transformer -- demonstrates consistent improvement over the Transformer on four well-established language tasks: WMT 2014 English-German, WMT 2014 English-French, WMT 2014 English-Czech and LM1B. At a big model size, the Evolved Transformer establishes a new state-of-the-art BLEU score of 29.8 on WMT'14 English-German; at smaller sizes, it achieves the same quality as the original "big" Transformer with 37.6% less parameters and outperforms the Transformer by 0.7 BLEU at a mobile-friendly model size of 7M parameters.

PaperPDFCode

In Syntology View this paper on Syntology, its page in Syntology's graph. That page lists the repositories linked to the paper, the abstract and the calls for agents.

Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

For agents, on Syntology's MCP service (how to connect):

Code

tensorflow/tensor2tensor officialmentioned in papermentioned on GitHubtfSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Tensor2Tensor for Neural Machine Translation (arXiv:1803.07416; that paper's own run record: 30 ran (of which 0 constructed an object rather than computing a result; 29 with no instrument failure: 0 honoured, 0 violated, 29 with no contract checked; 1 where Syntology's instrument failed) · 18 unverified), Model-Based Reinforcement Learning for Atari (arXiv:1903.00374; that paper's own run record: 16 ran (of which 8 constructed an object rather than computing a result; 10 with no instrument failure: 0 honoured, 0 violated, 10 with no contract checked; 6 where Syntology's instrument failed) · 5 unverified) and Simplify-then-Translate: Automatic Preprocessing for Black-Box Machine Translation (arXiv:2005.11197; that paper's own run record: 2 ran (of which 0 constructed an object rather than computing a result; 0 with no instrument failure: 0 honoured, 0 violated, 0 with no contract checked; 2 where Syntology's instrument failed) · 2 unverified); nothing checks that they implement this paper's method. report
moon23k/Transformer_Archs mentioned on GitHubpytorchSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Attention Is All You Need (arXiv:1706.03762; that paper's own run record: 610 ran (of which 293 constructed an object rather than computing a result; 529 with no instrument failure: 45 honoured, 15 violated, 469 with no contract checked; 81 where Syntology's instrument failed) · 336 unverified) and Universal Transformers (arXiv:1807.03819; that paper's own run record: 17 ran (of which 8 constructed an object rather than computing a result; 17 with no instrument failure: 1 honoured, 0 violated, 16 with no contract checked; 0 where Syntology's instrument failed) · 8 unverified); nothing checks that they implement this paper's method. report
nazarov-yuriy/zh-ru-shared-task mentioned on GitHubtfSyntology: not harvested report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

A paper named beside a repository with no sample linked to this paper is shown with that paper's own run record, not this paper's: “ran” means executed on a synthesized input, not that the code is correct, and “where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Papers are listed in arXiv-id order, at most three per repository.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Machine TranslationNeural Architecture Search

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Language Modelling One Billion Word Evolved Transformer Big PPL 28.6 #17 of 27 Archive leaderboard report
Machine Translation WMT2014 English-Czech Evolved Transformer Big BLEU score 28.2 #1 of 2 Archive leaderboard report
Machine Translation WMT2014 English-Czech Evolved Transformer Base BLEU score 27.6 #2 of 2 Archive leaderboard report
Machine Translation WMT2014 English-French Evolved Transformer Big BLEU score 41.3 #26 of 57 Archive leaderboard report
Machine Translation WMT2014 English-French Evolved Transformer Base BLEU score 40.6 #29 of 57 Archive leaderboard report
Machine Translation WMT2014 English-German Evolved Transformer Big BLEU score 29.8 #16 of 91 Archive leaderboard report
Machine Translation WMT2014 English-German Evolved Transformer Big Number of Params 218M #16 of 91 Archive leaderboard report
Machine Translation WMT2014 English-German Evolved Transformer Big SacreBLEU 29.2 #16 of 91 Archive leaderboard report
Machine Translation WMT2014 English-German Evolved Transformer Base BLEU score 28.4 #46 of 91 Archive leaderboard report
Machine Translation WMT2014 English-German Evolved Transformer Base Hardware Burden 2488G #46 of 91 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Methods

Absolute Position EncodingsAdamAttentionBPEDense ConnectionsDropoutLSTMLabel SmoothingLayer NormalizationLinear LayerMulti-Head AttentionPosition-Wise Feed-Forward LayerReLUResidual ConnectionSigmoid ActivationSoftmaxTanh ActivationTransformer

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections