Papers › Reformer: The Efficient Transformer

Reformer: The Efficient Transformer

13 Jan 2020ICLR 2020 1arXiv:2001.04451archive 2025-07-28

Nikita Kitaev, Łukasz Kaiser, Anselm Levskaya

Large Transformer models routinely achieve state-of-the-art results on a number of tasks but training these models can be prohibitively costly, especially on long sequences. We introduce two techniques to improve the efficiency of Transformers. For one, we replace dot-product attention by one that uses locality-sensitive hashing, changing its complexity from O(L²) to O(LlogL), where L is the length of the sequence. Furthermore, we use reversible residual layers instead of the standard residuals, which allows storing activations only once in the training process instead of N times, where N is the number of layers. The resulting model, the Reformer, performs on par with Transformer models while being much more memory-efficient and much faster on long sequences.

PaperPDFConference PDFCodeCode Syntology ranMethod as code

In Syntology View this paper on Syntology, its page in Syntology's graph. That page adds which of its functions are also present in other papers' code Syntology holds, with dates, and its dated run record as JSON and a README badge. It also 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

No sample from the archive's official repository (google/trax) is linked to this paper; the 8 samples below come from 1 community repository. Syntology harvested code from google/trax for another paper.

google/trax: Syntology's graph links these samples to Hierarchical Transformers Are More Efficient Language Models (arXiv:2110.13711; that paper's own run record: 3 ran (of which 0 constructed an object rather than computing a result; 3 with no instrument failure: 0 honoured, 2 violated, 1 with no contract checked; 0 where Syntology's instrument failed) · 1 unverified); nothing checks that they implement this paper's method.

Syntology 8 code samples harvested from 1 repository linked to this paper: 6 ran (of which 1 constructed an object rather than computing a result; 1 with no instrument failure: 0 honoured, 0 violated, 1 with no contract checked; 5 where Syntology's instrument failed) · 2 unverified. Where Syntology's instrument failed: 5 where our contract draft was wrong; failures of Syntology's instrument, not of the code.

By repository: community (archive-listed): 8 samples from 1 repository, 6 ran (1 with no instrument failure; 5 where Syntology's instrument failed). The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

No sample is linked to this paper from 5 of its repositories that Syntology harvested for other papers:

Levels written on 2026-09-22 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed.

google/trax officialmentioned on GitHubjaxSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Hierarchical Transformers Are More Efficient Language Models (arXiv:2110.13711; that paper's own run record: 3 ran (of which 0 constructed an object rather than computing a result; 3 with no instrument failure: 0 honoured, 2 violated, 1 with no contract checked; 0 where Syntology's instrument failed) · 1 unverified); nothing checks that they implement this paper's method. report
huggingface/transformers 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), HuggingFace's Transformers: State-of-the-art Natural Language Processing (arXiv:1910.03771; that paper's own run record: 3 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; 3 where Syntology's instrument failed) · 0 unverified), DeBERTa: Decoding-enhanced BERT with Disentangled Attention (arXiv:2006.03654; that paper's own run record: 4 ran (of which 0 constructed an object rather than computing a result; 3 with no instrument failure: 2 honoured, 1 violated, 0 with no contract checked; 1 where Syntology's instrument failed) · 9 unverified) and 1 more (every one in this page's JSON twin); nothing checks that they implement this paper's method. report
lucashueda/long_sentence_transformer mentioned on GitHubpytorchLGPL-3.0Syntology: not harvested report
sliao-mi-luku/Chatbot-Reformer mentioned on GitHubSyntology: not harvested report
sliao-mi-luku/NLP-Chatbot-Reformer-Trax mentioned on GitHubpytorchSyntology: not harvested report
MindCode-4/code-3 mindsporeSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to MathOdyssey: Benchmarking Mathematical Problem-Solving Skills in Large Language Models... (arXiv:2406.18321; that paper's own run record: 5 ran (of which 0 constructed an object rather than computing a result; 4 with no instrument failure: 0 honoured, 0 violated, 4 with no contract checked; 1 where Syntology's instrument failed) · 0 unverified); nothing checks that they implement this paper's method. report
Rick-McCoy/Reformer-pytorch pytorchSyntology: 8 samples, 6 ran (5 instrument-failed) · 2 unverified report
junnyu/paddle_reformer jaxSyntology: not harvested report
lucidrains/reformer-pytorch pytorchSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to RoFormer: Enhanced Transformer with Rotary Position Embedding (arXiv:2104.09864; that paper's own run record: 14 ran (of which 5 constructed an object rather than computing a result; 8 with no instrument failure: 1 honoured, 0 violated, 7 with no contract checked; 6 where Syntology's instrument failed) · 2 unverified) and Decentralized Attention Fails Centralized Signals: Rethinking Transformers for Medical... (arXiv:2602.18473; that paper's own run record: 12 ran (of which 6 constructed an object rather than computing a result; 8 with no instrument failure: 1 honoured, 0 violated, 7 with no contract checked; 4 where Syntology's instrument failed) · 4 unverified); nothing checks that they implement this paper's method. report
yangyucheng000/University mindsporeSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution (arXiv:2409.12191; that paper's own run record: 12 ran (of which 1 constructed an object rather than computing a result; 10 with no instrument failure: 4 honoured, 1 violated, 5 with no contract checked; 2 where Syntology's instrument failed) · 0 unverified); nothing checks that they implement this paper's method. 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

8 samples harvested: 6 ran (of which 1 constructed an object rather than computing a result; 1 with no instrument failure: 0 honoured, 0 violated, 1 with no contract checked; 5 where Syntology's instrument failed) · 2 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page has not changed since , the first build that kept a record date for it; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

5ran · our draft was wrong
1ran
2unverified

Licence: 0 of the 8 samples are pointer only, meaning Syntology does not serve that copy's text. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from Rick-McCoy/Reformer-pytorch. “Ran” means the sample executed on a synthesized input. It does not mean the output is correct, and nothing here reproduces the paper's results. A line beginning “fixture:” under a sample that ran gives the arguments Syntology's fixture called the sample with, each as name:type/rank/element kind (? where none was recorded), and the type of what the fixture's run returned, exactly as the harness recorded them; the arguments and the returned type were recorded by two separate runs of the same fixture. No output value is shown, and a returned type says nothing about whether the output was right. “Level recorded” gives the UTC date when Syntology's graph wrote the sample's verification level, not when the code executed; where the level was written before Syntology's graph kept that date, the row says “level recorded, date not recorded”. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself. “Our draft was wrong” and “the contract check could not run” are failures of Syntology's instrument, not of the code: the sample ran on Syntology's synthesized input, and then the drafted contract did not describe it, or the check itself raised an error (RAISES), lacked a dependency (DEP_MISSING) or ran out of time (TIMEOUT); hover the label for which. A metamorphic tier, where shown, is evidence of the output's shape, not that the sample implements the method.

Each sample ends with its code_sha256. A sample's code_sha256 is Syntology's identity for that exact code. It is printed in full (16 hexadecimal characters for every sample in this build), and get_code(code_sha256="…") takes it exactly as printed. An agent calls it on Syntology's MCP service (how to connect); click an identity to copy that call.

Repository labels, per sample. official repository: The archive marks this repository official for the paper. named in the paper: The archive records that the paper mentions this repository; it is not marked official. community (archive-listed): In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper. found in paper text by Syntology: Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted. community: Not in the archive's code links for this paper; a community repository Syntology harvested. Samples are grouped by repository: a repository marked official first, then by how directly the repository is tied to the paper, then in the archive's order; never by how many ran. Licence labels name the repository's licence as recorded at harvest. “Pointer only” means Syntology does not serve that copy's text, for one of six reasons: no licence file was found; the licence was not identified; the licence is recorded as permissive but that copy's record is not marked cleared; the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar); this paper's copy was not located, so its licence is not recorded; or this paper's copy is cleared, but the stored copy of the same code that Syntology serves is not. Some licences outside that list permit redistribution, such as WTFPL, and GPL-3.0 under its conditions; they are simply not on the list. Hover a licence label for the reason. File links open the file on GitHub at the default branch, which may have changed since the harvest.

A function or class name links to its page when the name occurs in the harvested code of at least 5 papers; the name is a string, not an identity.

Rick-McCoy/Reformer-pytorch community (archive-listed) 8 samples, 6 ran (5 instrument-failed) · 2 unverified
LocalitySensitiveHash model/attention.py ran MIT (permissive) · f730a019cd6dfa68 · report fixture: returned LocalitySensitiveHash · level recorded 2026-09-22
expand in 29 papers model/attention.py ran · our draft was wrong fingerprinted MIT (permissive) · bde1d7f419f43bee · report fixture: signature input_tensor:Tensor/2/float, dim:int/0/int, num:int/0/int · returned Tensor · level recorded 2026-09-22
expand_gather model/attention.py ran · our draft was wrong MIT (permissive) · e25bbd4d3edcecd5 · report fixture: signature input_tensor:Tensor/3/float, dim:int/0/int, index:Tensor/2/int, expand_dim:int/0/int, num:int/0/int · returned Tensor · level recorded 2026-09-22
get_dup_keys model/attention.py ran · our draft was wrong fingerprinted MIT (permissive) · fa28de4db6c8dda3 · report fixture: signature input_tensor:Tensor/3/int, rounds:int/0/int · returned Tensor · level recorded 2026-09-22
look_back model/attention.py ran · our draft was wrong fingerprinted MIT (permissive) · 48898f78f7058226 · report fixture: signature input_tensor:Tensor/3/float · returned Tensor · level recorded 2026-09-22
reverse_sort model/attention.py ran · our draft was wrong MIT (permissive) · 4afdf83df6c27154 · report fixture: signature indice:Tensor/2/int, dim:int/0/int · returned Tensor · level recorded 2026-09-22
LSHAttention model/attention.py unverified MIT (permissive) · 6874256369242fb8 · report
deterministic_dropout model/attention.py unverified MIT (permissive) · 939cb0f4831afed5 · report

Method as code Syntology

What Syntology's graph records as this paper's claimed method, lifted into code: each assembly names its bricks (functions and classes), each with its own role and its own verification level. An assembly is not a check that the code implements the method correctly or reproduces the paper's results. Levels are each brick's own verification level in Syntology's graph (level 0: no recorded run), listed brick by brick and never combined into one level; “ran” means the brick executed on a synthesized input, not that its output is correct, and the instrument figure counts failures of Syntology's instrument, not of the code. Where a model read the code against the paper's claim, each assembly's label verdict is one model's judgement, kept in this page's JSON twin; it is not calibrated, so no card prints it.

1 assembly on this paper: 1 harvested, 0 generated.

harvested Assembly for REFORMER (graph method: REFORMER). 8 bricks; individually on a fixture: 6 ran (of which 1 constructed an object rather than computing a result; 1 with no instrument failure: 0 honoured, 0 violated, 1 with no contract checked; 5 where Syntology's instrument failed) · 2 unverified. The assembly has not been run end-to-end.

  1. LSHAttention entry · level 0 · unverified · 6874256369242fb8
  2. LocalitySensitiveHash helper · level 1 · ran · f730a019cd6dfa68
  3. deterministic_dropout helper · level 0 · unverified · 939cb0f4831afed5
  4. expand helper · level 1 · ran · our draft was wrong · bde1d7f419f43bee
  5. expand_gather helper · level 1 · ran · our draft was wrong · e25bbd4d3edcecd5
  6. get_dup_keys helper · level 1 · ran · our draft was wrong · fa28de4db6c8dda3
  7. look_back helper · level 1 · ran · our draft was wrong · 48898f78f7058226
  8. reverse_sort helper · level 1 · ran · our draft was wrong · 4afdf83df6c27154

Each brick ends with its code_sha256, Syntology's identity for that exact code; click one to copy the get_code call that returns it. For agents, on Syntology's MCP service: assembly(arxiv_id="2001.04451") returns these assemblies with each brick's role and wiring (how to connect).

Tasks

D4RLImage GenerationLanguage ModellingOffline RLOpen-Domain Question AnsweringQuestion Answering

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
D4RL D4RL Reformer Average Reward 63.9 #8 of 9 Archive leaderboard report
Image Generation ImageNet 64x64 Reformer (12 layers) Bits per dim 3.710 #56 of 65 Archive leaderboard report
Image Generation ImageNet 64x64 Reformer (6 layers) Bits per dim 3.740 #58 of 65 Archive leaderboard report
Language Modelling WikiText-103 Reformer 125M Test perplexity 26.0 #62 of 89 Archive leaderboard report
Open-Domain Question Answering SearchQA Locality-Sensitive Hashing EM 66.0 #2 of 14 Archive leaderboard report
Question Answering Natural Questions (long) Locality-Sensitive Hashing F1 75.5 #3 of 13 Archive leaderboard report
Question Answering Quasart-T Locality-Sensitive Hashing EM 53.2 #2 of 7 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

Introduced by this paper: Reformer

Absolute Position EncodingsAdafactorAdamAttentionBPEDense ConnectionsDropoutLSH AttentionLabel SmoothingLayer NormalizationLinear LayerMulti-Head AttentionPosition-Wise Feed-Forward LayerReLUReformerResidual ConnectionReversible Residual BlockSentencePieceSoftmaxTransformer

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