Papers › Bottleneck Transformers for Visual Recognition

Bottleneck Transformers for Visual Recognition

27 Jan 2021CVPR 2021 1arXiv:2101.11605archive 2025-07-28

Aravind Srinivas, Tsung-Yi Lin, Niki Parmar, Jonathon Shlens, Pieter Abbeel, Ashish Vaswani

We present BoTNet, a conceptually simple yet powerful backbone architecture that incorporates self-attention for multiple computer vision tasks including image classification, object detection and instance segmentation. By just replacing the spatial convolutions with global self-attention in the final three bottleneck blocks of a ResNet and no other changes, our approach improves upon the baselines significantly on instance segmentation and object detection while also reducing the parameters, with minimal overhead in latency. Through the design of BoTNet, we also point out how ResNet bottleneck blocks with self-attention can be viewed as Transformer blocks. Without any bells and whistles, BoTNet achieves 44.4% Mask AP and 49.7% Box AP on the COCO Instance Segmentation benchmark using the Mask R-CNN framework; surpassing the previous best published single model and single scale results of ResNeSt evaluated on the COCO validation set. Finally, we present a simple adaptation of the BoTNet design for image classification, resulting in models that achieve a strong performance of 84.7% top-1 accuracy on the ImageNet benchmark while being up to 1.64x faster in compute time than the popular EfficientNet models on TPU-v3 hardware. We hope our simple and effective approach will serve as a strong baseline for future research in self-attention models for vision

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

Syntology 49 code samples harvested from 9 repositories linked to this paper: 26 ran (of which 9 constructed an object rather than computing a result; 19 with no instrument failure: 1 honoured, 0 violated, 18 with no contract checked; 7 where Syntology's instrument failed) · 23 unverified. Where Syntology's instrument failed: 2 where our contract draft was wrong; 5 where the contract check could not run; failures of Syntology's instrument, not of the code.

By repository: community (archive-listed): 49 samples from 9 repositories, 26 ran (19 with no instrument failure; 7 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 3 of its repositories that Syntology harvested for other papers:

Levels written between 2026-09-22 and 2026-09-24 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed.

13 repositories listed; official and paper-mentioned ones first.

BIGBALLON/distribuuuu mentioned on GitHubpytorchSyntology: 7 samples, 7 ran (3 instrument-failed) · 0 unverified report
BR-IDL/PaddleViT mentioned on GitHubpaddleApache-2.0Syntology: not harvested report
CandiceD17/Bottleneck-Transformers-for-Visual-Recognition mentioned on GitHubpytorchSyntology: 4 samples, 2 ran (0 instrument-failed) · 2 unverified report
MartinGer/Bottleneck-Transformers-for-Visual-Recognition mentioned on GitHubpytorchSyntology: 1 sample, 1 ran (0 instrument-failed) · 0 unverified report
NZ99/bottleneck-transformer-flax mentioned on GitHubjaxSyntology: 7 samples, 0 ran · 7 unverified report
kevin-ssy/ViP mentioned on GitHubpytorchMITSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Scaling Local Self-Attention for Parameter Efficient Visual Backbones (arXiv:2103.12731; that paper's own run record: 18 ran (of which 4 constructed an object rather than computing a result; 11 with no instrument failure: 1 honoured, 0 violated, 10 with no contract checked; 7 where Syntology's instrument failed) · 2 unverified); nothing checks that they implement this paper's method. report
leaderj1001/BottleneckTransformers mentioned on GitHubpytorchSyntology: 2 samples, 2 ran (0 instrument-failed) · 0 unverified report
lucidrains/bottleneck-transformer-pytorch mentioned on GitHubpytorchSyntology: 7 samples, 4 ran (0 instrument-failed) · 3 unverified report
nachiket273/VisTrans mentioned on GitHubpytorchSyntology: 11 samples, 2 ran (1 instrument-failed) · 9 unverified report
rwightman/pytorch-image-models mentioned on GitHubpytorchSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to AdamP: Slowing Down the Slowdown for Momentum Optimizers on Scale-invariant Weights (arXiv:2006.08217; that paper's own run record: 4 ran (of which 1 constructed an object rather than computing a result; 2 with no instrument failure: 1 honoured, 0 violated, 1 with no contract checked; 2 where Syntology's instrument failed) · 0 unverified); nothing checks that they implement this paper's method. report
scheshmi/BottleneckTransformers mentioned on GitHubpytorchSyntology: 3 samples, 2 ran (0 instrument-failed) · 1 unverified report
The-AI-Summer/self_attention pytorchSyntology: 7 samples, 6 ran (3 instrument-failed) · 1 unverified report
leondgarse/keras_cv_attention_models tfSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to RepViT: Revisiting Mobile CNN From ViT Perspective (arXiv:2307.09283; that paper's own run record: 6 ran (of which 0 constructed an object rather than computing a result; 3 with no instrument failure: 0 honoured, 0 violated, 3 with no contract checked; 3 where Syntology's instrument failed) · 6 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

49 samples harvested: 26 ran (of which 9 constructed an object rather than computing a result; 19 with no instrument failure: 1 honoured, 0 violated, 18 with no contract checked; 7 where Syntology's instrument failed) · 23 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.

1ran · honoured contract
2ran · our draft was wrong
5ran; the contract check could not run
18ran
23unverified

Licence: 8 of the 49 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 9 repositories linked to this paper, official or community; the list below is grouped by repository, each group headed by its name and what it is to the paper. “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”. For 1 sample that ran, the graph records a signature this page does not show: for a class the fixture built without calling, the harness called the object with inputs it made up itself, so the recorded signature is the harness's probe, not the fixture's arguments. “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. A census tag appears when Syntology's census ran implementations with that name on one shared input per bucket: it gives how many were compared, from how many papers, and how many distinct outputs they produced, and names this sample's bucket when this sample was one of them.

BIGBALLON/distribuuuu community (archive-listed) 7 samples, 7 ran (3 instrument-failed) · 0 unverified
AbsPosEmb distribuuuu/models/botnet.py ran MIT (permissive) · bf90d53646507d1c · report fixture: returned AbsPosEmb · level recorded 2026-09-22
BoTBlock distribuuuu/models/botnet.py ran MIT (permissive) · 3f06d18fc36f803d · report fixture: returned BoTBlock · level recorded 2026-09-22
MHSA distribuuuu/models/botnet.py ran MIT (permissive) · 3d387b9ea3a491be · report fixture: returned MHSA · level recorded 2026-09-22
RelPosEmb distribuuuu/models/botnet.py ran MIT (permissive) · 1e8e321a97002a33 · report fixture: returned RelPosEmb · level recorded 2026-09-22
expand_dim in 11 papers
census: 3 implementations from 5 papers, 1 distinct output on one shared input this sample: bucket 1 of 1, arg 1: rank 2, kind float, dtype float32
distribuuuu/models/botnet.py ran · our draft was wrong fingerprinted MIT (permissive) · ddac376537265a32 · report fixture: signature t:Tensor/2/float, dim:int/0/int, k:int/0/int · returned Tensor · level recorded 2026-09-22
rel_to_abs distribuuuu/models/botnet.py ran; the contract check could not run (RAISES) MIT (permissive) · 560641da841eeb15 · report fixture: returned Tensor · level recorded 2026-09-22
relative_logits_1d distribuuuu/models/botnet.py ran; the contract check could not run (RAISES) MIT (permissive) · 93ec7f5a59863d1a · report fixture: returned Tensor · level recorded 2026-09-22
CandiceD17/Bottleneck-Transformers-for-Visual-Recognition community (archive-listed) 4 samples, 2 ran (0 instrument-failed) · 2 unverified
AbsPosEmb model/botnet.py ran · metamorphic tier: invariant no licence file found · pointer only · abdfbeb02321716b · report fixture: signature q:Tensor/4/float · returned Tensor · level recorded 2026-09-22
RelPosEmb model/botnet.py ran no licence file found · pointer only · 90d22099bc8bc6ad · report fixture: signature q:Tensor/4/float · returned Tensor · level recorded 2026-09-22
BoTBlock model/botnet.py unverified no licence file found · pointer only · c89f1a9517a0905e · report
MHSA model/botnet.py unverified no licence file found · pointer only · 0eb9211a00bd7524 · report
MartinGer/Bottleneck-Transformers-for-Visual-Recognition community (archive-listed) 1 sample, 1 ran (0 instrument-failed) · 0 unverified
BoTNet_Layer BoTNet_Layer.py ran no licence file found · pointer only · eda28b8215e9ff22 · report fixture: returned BoTNet_Layer · level recorded 2026-09-22
NZ99/bottleneck-transformer-flax community (archive-listed) 7 samples, 0 ran · 7 unverified
BoTBlock bottleneck_transformer_flax/botnet.py unverified MIT (permissive) · 635303b35edfe8e2 · report
BoTMHSA bottleneck_transformer_flax/botnet.py unverified MIT (permissive) · 4246cb32d13d3b98 · report
BoTNet bottleneck_transformer_flax/botnet.py unverified MIT (permissive) · 93c353b1b05ef701 · report
BoTNetConfig bottleneck_transformer_flax/botnet.py unverified MIT (permissive) · 58200aabf211669a · report
BottleneckResNetBlock bottleneck_transformer_flax/botnet.py unverified MIT (permissive) · 685fb6581dbbfbec · report
RelativeLogits bottleneck_transformer_flax/botnet.py unverified MIT (permissive) · c92dd55c1dc2fa04 · report
SELayer in 12 papers bottleneck_transformer_flax/botnet.py unverified MIT (permissive) · 09e1a6a94cc17376 · report
leaderj1001/BottleneckTransformers community (archive-listed) 2 samples, 2 ran (0 instrument-failed) · 0 unverified
Bottleneck in 98 papers model.py ran MIT (permissive) · a1c9f8ce96f754a0 · report fixture: signature x:Tensor/4/float · returned Tensor · level recorded 2026-09-24
MHSA model.py ran · metamorphic tier: invariant MIT (permissive) · b4f23ec7b5e89157 · report fixture: signature x:Tensor/4/float · returned Tensor · level recorded 2026-09-22
lucidrains/bottleneck-transformer-pytorch community (archive-listed) 7 samples, 4 ran (0 instrument-failed) · 3 unverified
AbsPosEmb bottleneck_transformer_pytorch/bottleneck_transformer_pytorch.py ran · metamorphic tier: invariant MIT (permissive) · 55cdbdffabd748b8 · report fixture: signature q:Tensor/4/float · returned Tensor · level recorded 2026-09-22
Attention in 344 papers
census: 13 implementations from 12 papers, 10 distinct outputs across 4 buckets, one shared input per bucket
bottleneck_transformer_pytorch/bottleneck_transformer_pytorch.py ran · metamorphic tier: invariant MIT (permissive) · 1ee91139c3f9ceb2 · report fixture: signature fmap:Tensor/4/float · returned Tensor · level recorded 2026-09-22
RelPosEmb bottleneck_transformer_pytorch/bottleneck_transformer_pytorch.py ran · metamorphic tier: invariant MIT (permissive) · 5f7d1c16442b0f3c · report fixture: signature q:Tensor/4/float · returned Tensor · level recorded 2026-09-22
pair in 101 papers bottleneck_transformer_pytorch/bottleneck_transformer_pytorch.py ran · honoured contract fingerprinted MIT (permissive) · 3e27684df29ad41d · report fixture: signature x:Tensor/2/float · returned tuple · level recorded 2026-09-22
BottleBlock bottleneck_transformer_pytorch/bottleneck_transformer_pytorch.py unverified MIT (permissive) · 957f9748a5caf504 · report
rel_to_abs bottleneck_transformer_pytorch/bottleneck_transformer_pytorch.py unverified MIT (permissive) · 2c273760ec522435 · report
relative_logits_1d bottleneck_transformer_pytorch/bottleneck_transformer_pytorch.py unverified MIT (permissive) · 5b41b7a682f4c678 · report
nachiket273/VisTrans community (archive-listed) 11 samples, 2 ran (1 instrument-failed) · 9 unverified
GroupPointWise vistrans/models/botnet.py ran fingerprinted MIT (permissive) · 228f55a9b608179b · report fixture: returned GroupPointWise · level recorded 2026-09-22
relative_logits vistrans/models/botnet.py ran; the contract check could not run (RAISES) fingerprinted MIT (permissive) · bb969778cf283586 · report fixture: signature w:Tensor/2/float, h:Tensor/2/float, q:Tensor/5/float · returned Tensor · level recorded 2026-09-22
AbsolutePosEmb vistrans/models/botnet.py unverified MIT (permissive) · 60b38ccc9cd86edd · report
BotNet vistrans/models/botnet.py unverified MIT (permissive) · 771d64f60cf10b0d · report
BotnetX vistrans/models/botnet.py unverified MIT (permissive) · 3d7d4980902b0da3 · report
Bottleneck in 98 papers vistrans/models/botnet.py unverified MIT (permissive) · 6fd5faa3cfeccf67 · report
MHSA vistrans/models/botnet.py unverified MIT (permissive) · 4a8e8eb7d2edff3d · report
RelativePosEmb vistrans/models/botnet.py unverified MIT (permissive) · e1685f603e910d5a · report
absolute_logits vistrans/models/botnet.py unverified MIT (permissive) · 8c47eb5f2d30130b · report
rel_to_abs vistrans/models/botnet.py unverified MIT (permissive) · 1d5fc659fc20f4b8 · report
relative_logits_1d vistrans/models/botnet.py unverified MIT (permissive) · dde3921cad5d0cc0 · report
scheshmi/BottleneckTransformers community (archive-listed) 3 samples, 2 ran (0 instrument-failed) · 1 unverified
Block in 252 papers
census: 5 implementations from 5 papers, 3 distinct outputs across 2 buckets, one shared input per bucket
model.py ran no licence file found · pointer only · 5f213d496b96f84d · report fixture: signature x:Tensor/4/float · returned Tensor · level recorded 2026-09-24
MHSA model.py ran · metamorphic tier: invariant fingerprinted no licence file found · pointer only · e3afa08cab48c4cf · report fixture: signature x:Tensor/4/float · returned Tensor · level recorded 2026-09-22
BoTNet model.py unverified no licence file found · pointer only · 14a63975366ccc24 · report
The-AI-Summer/self_attention community (archive-listed) 7 samples, 6 ran (3 instrument-failed) · 1 unverified
BottleneckAttention self_attention_cv/bottleneck_transformer/bot_block.py ran · metamorphic tier: deterministic MIT (permissive) · 47c97b7ed35a625f · report fixture: returned BottleneckAttention · level recorded 2026-09-22
RelPosEmb1D self_attention_cv/bottleneck_transformer/bot_block.py ran · metamorphic tier: deterministic MIT (permissive) · 612b904217f4544b · report fixture: returned RelPosEmb1D · level recorded 2026-09-22
RelPosEmb2D self_attention_cv/bottleneck_transformer/bot_block.py ran · metamorphic tier: deterministic MIT (permissive) · 22fb48daf7ae0cc4 · report fixture: returned RelPosEmb2D · level recorded 2026-09-22
_conv2d1x1 self_attention_cv/bottleneck_transformer/bot_block.py ran · our draft was wrong MIT (permissive) · b4c16d4e84a73b3d · report fixture: signature in_channels:int/0/int, out_channels:int/0/int, stride:int/0/int · returned Sequential · level recorded 2026-09-22
rel_pos_emb_1d self_attention_cv/bottleneck_transformer/bot_block.py ran; the contract check could not run (RAISES) · metamorphic tier: invariant fingerprinted MIT (permissive) · 227b65e34b12edd5 · report fixture: signature q:Tensor/4/float, rel_emb:Tensor/2/float, shared_heads:bool/0/bool · returned Tensor · level recorded 2026-09-22
relative_to_absolute self_attention_cv/bottleneck_transformer/bot_block.py ran; the contract check could not run (RAISES) · metamorphic tier: invariant fingerprinted MIT (permissive) · c692e058b3b543f2 · report fixture: signature q:Tensor/4/float · returned Tensor · level recorded 2026-09-22
BottleneckBlock in 5 papers self_attention_cv/bottleneck_transformer/bot_block.py unverified MIT (permissive) · 31358a487667e5f1 · 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. Siblings are assemblies lifted from this paper in the same batch for the same method: alternative readings kept side by side; none replaces or supersedes another.

9 assemblies on this paper: 9 harvested, 0 generated, listed in 1 groups (siblings lifted in one batch form one group).

9 sibling assemblies for BoTNet (harvested), lifted in the same batch on 2026-09-09: alternative readings kept side by side; none replaces or supersedes another.

harvested Assembly for BoTNet (graph method: BoTNet). 7 bricks; individually on a fixture: 6 ran (of which 3 constructed an object rather than computing a result; 3 with no instrument failure: 0 honoured, 0 violated, 3 with no contract checked; 3 where Syntology's instrument failed) · 1 unverified. The assembly has not been run end-to-end.

one of 9 siblings

  1. BottleneckBlock entry · level 0 · unverified · 31358a487667e5f1
  2. BottleneckAttention model · level 1 · ran · 47c97b7ed35a625f
  3. _conv2d1x1 helper · level 1 · ran · our draft was wrong · b4c16d4e84a73b3d
  4. RelPosEmb2D helper · level 1 · ran · 22fb48daf7ae0cc4
  5. RelPosEmb1D helper · level 1 · ran · 612b904217f4544b
  6. rel_pos_emb_1d helper · level 1 · ran; the contract check could not run (RAISES) · 227b65e34b12edd5
  7. relative_to_absolute helper · level 1 · ran; the contract check could not run (RAISES) · c692e058b3b543f2

harvested Assembly for BoTNet (graph method: BoTNet). 12 bricks; individually on a fixture: 2 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; 1 where Syntology's instrument failed) · 10 unverified. The assembly has not been run end-to-end.

one of 9 siblings

  1. BotNet entry · level 0 · unverified · 771d64f60cf10b0d
  2. BotnetX model · level 0 · unverified · 3d7d4980902b0da3
  3. MHSA model · level 0 · unverified · 4a8e8eb7d2edff3d
  4. Bottleneck model · level 0 · unverified · 6fd5faa3cfeccf67
  5. AbsolutePosEmb helper · level 0 · unverified · 60b38ccc9cd86edd
  6. GroupPointWise helper · level 1 · ran · 228f55a9b608179b
  7. RelativePosEmb helper · level 0 · unverified · e1685f603e910d5a
  8. absolute_logits helper · level 0 · unverified · 8c47eb5f2d30130b
  9. ntuple helper · level 0 · unverified · 7b701d0ab4c91bf0
  10. relative_logits helper · level 1 · ran; the contract check could not run (RAISES) · bb969778cf283586
  11. relative_logits_1d helper · level 0 · unverified · dde3921cad5d0cc0
  12. rel_to_abs helper · level 0 · unverified · 1d5fc659fc20f4b8

harvested Assembly for BoTNet (graph method: BoTNet). 1 brick; individually on a fixture: 1 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; 0 where Syntology's instrument failed) · 0 unverified; the one sample that ran constructed an object rather than computing a result. The assembly has not been run end-to-end.

one of 9 siblings

  1. BoTNet_Layer entry · level 1 · ran · eda28b8215e9ff22

harvested Assembly for BoTNet (graph method: BoTNet). 7 bricks; individually on a fixture: 7 ran (of which 4 constructed an object rather than computing a result; 4 with no instrument failure: 0 honoured, 0 violated, 4 with no contract checked; 3 where Syntology's instrument failed) · 0 unverified. The assembly has not been run end-to-end.

one of 9 siblings

  1. BoTBlock entry · level 1 · ran · 3f06d18fc36f803d
  2. MHSA model · level 1 · ran · 3d387b9ea3a491be
  3. AbsPosEmb model · level 1 · ran · bf90d53646507d1c
  4. RelPosEmb model · level 1 · ran · 1e8e321a97002a33
  5. relative_logits_1d helper · level 1 · ran; the contract check could not run (RAISES) · 93ec7f5a59863d1a
  6. expand_dim helper · level 1 · ran · our draft was wrong · ddac376537265a32
  7. rel_to_abs helper · level 1 · ran; the contract check could not run (RAISES) · 560641da841eeb15

harvested Assembly for BoTNet (graph method: BoTNet). 7 bricks; individually on a fixture: 5 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 0 honoured, 0 violated, 2 with no contract checked; 3 where Syntology's instrument failed) · 2 unverified. The assembly has not been run end-to-end.

one of 9 siblings

  1. BoTBlock entry · level 0 · unverified · c89f1a9517a0905e
  2. MHSA model · level 0 · unverified · 0eb9211a00bd7524
  3. AbsPosEmb model · level 1 · ran · abdfbeb02321716b
  4. RelPosEmb model · level 1 · ran · 90d22099bc8bc6ad
  5. relative_logits_1d helper · level 1 · ran; the contract check could not run (RAISES) · 93ec7f5a59863d1a
  6. expand_dim helper · level 1 · ran · our draft was wrong · ddac376537265a32
  7. rel_to_abs helper · level 1 · ran; the contract check could not run (RAISES) · 560641da841eeb15

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

one of 9 siblings

  1. Bottleneck entry · level 1 · ran · a1c9f8ce96f754a0
  2. MHSA model · level 1 · ran · b4f23ec7b5e89157

harvested Assembly for BoTNet (graph method: BoTNet). 3 bricks; individually on a fixture: 2 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 0 honoured, 0 violated, 2 with no contract checked; 0 where Syntology's instrument failed) · 1 unverified. The assembly has not been run end-to-end.

one of 9 siblings

  1. BoTNet entry · level 0 · unverified · 14a63975366ccc24
  2. Block model · level 1 · ran · 5f213d496b96f84d
  3. MHSA model · level 1 · ran · e3afa08cab48c4cf

harvested Assembly for BoTNet (graph method: BoTNet). 8 bricks; individually on a fixture: 5 ran (of which 0 constructed an object rather than computing a result; 4 with no instrument failure: 1 honoured, 0 violated, 3 with no contract checked; 1 where Syntology's instrument failed) · 3 unverified. The assembly has not been run end-to-end.

one of 9 siblings

  1. BottleBlock entry · level 0 · unverified · 957f9748a5caf504
  2. Attention model · level 1 · ran · 1ee91139c3f9ceb2
  3. AbsPosEmb model · level 1 · ran · 55cdbdffabd748b8
  4. RelPosEmb model · level 1 · ran · 5f7d1c16442b0f3c
  5. pair helper · level 1 · ran · honoured contract · 3e27684df29ad41d
  6. relative_logits_1d helper · level 0 · unverified · 5b41b7a682f4c678
  7. expand_dim helper · level 1 · ran · our draft was wrong · 042da966a79899bf
  8. rel_to_abs helper · level 0 · unverified · 2c273760ec522435

harvested Assembly for BoTNet (graph method: BoTNet). 7 bricks; individually on a fixture: 0 ran · 7 unverified. The assembly has not been run end-to-end.

one of 9 siblings

  1. BoTNet entry · level 0 · unverified · 93c353b1b05ef701
  2. BoTBlock model · level 0 · unverified · 635303b35edfe8e2
  3. BoTNetConfig helper · level 0 · unverified · 58200aabf211669a
  4. BottleneckResNetBlock model · level 0 · unverified · 685fb6581dbbfbec
  5. BoTMHSA model · level 0 · unverified · 4246cb32d13d3b98
  6. SELayer model · level 0 · unverified · 09e1a6a94cc17376
  7. RelativeLogits model · level 0 · unverified · c92dd55c1dc2fa04

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="2101.11605") returns these assemblies with each brick's role and wiring (how to connect).

Tasks

Image ClassificationInstance SegmentationObject DetectionSegmentationimage-classificationobject-detection

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Image Classification ImageNet BoTNet T7 Number of params 75.1M #298 of 1060 Archive leaderboard report
Image Classification ImageNet BoTNet T7 Top 1 Accuracy 84.7% #298 of 1060 Archive leaderboard report
Image Classification ImageNet BoTNet T7-320 Top 1 Accuracy 84.2% #333 of 1060 Archive leaderboard report
Image Classification ImageNet BoTNet T6 Number of params 53.9M #369 of 1060 Archive leaderboard report
Image Classification ImageNet BoTNet T6 Top 1 Accuracy 84% #369 of 1060 Archive leaderboard report
Image Classification ImageNet SENet-350 Top 1 Accuracy 83.8% #385 of 1060 Archive leaderboard report
Image Classification ImageNet BoTNet T5 GFLOPs 19.3 #421 of 1060 Archive leaderboard report
Image Classification ImageNet BoTNet T5 Top 1 Accuracy 83.5% #421 of 1060 Archive leaderboard report
Image Classification ImageNet BoTNet T4 GFLOPs 10.9 #504 of 1060 Archive leaderboard report
Image Classification ImageNet BoTNet T4 Number of params 54.7M #504 of 1060 Archive leaderboard report
Image Classification ImageNet BoTNet T4 Top 1 Accuracy 82.8% #504 of 1060 Archive leaderboard report
Image Classification ImageNet SENet-152 Number of params 66.6M #568 of 1060 Archive leaderboard report
Image Classification ImageNet SENet-152 Top 1 Accuracy 82.2% #568 of 1060 Archive leaderboard report
Image Classification ImageNet BoTNet T3 GFLOPs 7.3 #619 of 1060 Archive leaderboard report
Image Classification ImageNet BoTNet T3 Number of params 33.5M #619 of 1060 Archive leaderboard report
Image Classification ImageNet BoTNet T3 Top 1 Accuracy 81.7% #619 of 1060 Archive leaderboard report
Image Classification ImageNet SENet-101 Number of params 49.2M #645 of 1060 Archive leaderboard report
Image Classification ImageNet SENet-101 Top 1 Accuracy 81.4% #645 of 1060 Archive leaderboard report
Image Classification ImageNet ResNet-101 Number of params 44.4M #727 of 1060 Archive leaderboard report
Image Classification ImageNet ResNet-101 Top 1 Accuracy 80% #727 of 1060 Archive leaderboard report
Image Classification ImageNet SENet-50 Number of params 28.02M #762 of 1060 Archive leaderboard report
Image Classification ImageNet SENet-50 Top 1 Accuracy 79.4% #762 of 1060 Archive leaderboard report
Image Classification ImageNet ResNet-50 Number of params 25.5M #805 of 1060 Archive leaderboard report
Image Classification ImageNet ResNet-50 Top 1 Accuracy 78.8% #805 of 1060 Archive leaderboard report
Instance Segmentation COCO minival BoTNet 200 (Mask R-CNN, single scale, 72 epochs) mask AP 44.4 #55 of 93 Archive leaderboard report
Instance Segmentation COCO minival BoTNet 152 (Mask R-CNN, single scale, 72 epochs) mask AP 43.7 #59 of 93 Archive leaderboard report
Instance Segmentation COCO minival BoTNet 50 (72 epochs) mask AP 40.7 #71 of 93 Archive leaderboard report
Object Detection COCO minival BoTNet 200 (Mask R-CNN, single scale, 72 epochs) AP50 71.3 #83 of 220 Archive leaderboard report
Object Detection COCO minival BoTNet 200 (Mask R-CNN, single scale, 72 epochs) AP75 54.6 #83 of 220 Archive leaderboard report
Object Detection COCO minival BoTNet 200 (Mask R-CNN, single scale, 72 epochs) box AP 49.7 #83 of 220 Archive leaderboard report
Object Detection COCO minival BoTNet 152 (Mask R-CNN, single scale, 72 epochs) AP50 71 #84 of 220 Archive leaderboard report
Object Detection COCO minival BoTNet 152 (Mask R-CNN, single scale, 72 epochs) AP75 54.2 #84 of 220 Archive leaderboard report
Object Detection COCO minival BoTNet 152 (Mask R-CNN, single scale, 72 epochs) box AP 49.5 #84 of 220 Archive leaderboard report
Object Detection COCO minival BoTNet 50 (72 epochs) box AP 45.9 #111 of 220 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: Bottleneck Transformer Block

1x1 ConvolutionAttentionAverage PoolingBatch NormalizationBottleneck TransformerBottleneck Transformer BlockConvolutionCosine AnnealingDense ConnectionsLabel SmoothingMask R-CNNMax PoolingPointwise ConvolutionRPNRandAugmentRandom Resized CropReLUResNeStResidual ConnectionRoIAlignSGD with MomentumSiLUSigmoid ActivationSoftmaxSplit AttentionSqueeze-and-Excitation BlockWeight Decay

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