Papers › Unified Visual Transformer Compression

Unified Visual Transformer Compression

15 Mar 2022ICLR 2022 4arXiv:2203.08243archive 2025-07-28

Shixing Yu, Tianlong Chen, Jiayi Shen, Huan Yuan, Jianchao Tan, Sen yang, Ji Liu, Zhangyang Wang

Vision transformers (ViTs) have gained popularity recently. Even without customized image operators such as convolutions, ViTs can yield competitive performance when properly trained on massive data. However, the computational overhead of ViTs remains prohibitive, due to stacking multi-head self-attention modules and else. Compared to the vast literature and prevailing success in compressing convolutional neural networks, the study of Vision Transformer compression has also just emerged, and existing works focused on one or two aspects of compression. This paper proposes a unified ViT compression framework that seamlessly assembles three effective techniques: pruning, layer skipping, and knowledge distillation. We formulate a budget-constrained, end-to-end optimization framework, targeting jointly learning model weights, layer-wise pruning ratios/masks, and skip configurations, under a distillation loss. The optimization problem is then solved using the primal-dual algorithm. Experiments are conducted with several ViT variants, e.g. DeiT and T2T-ViT backbones on the ImageNet dataset, and our approach consistently outperforms recent competitors. For example, DeiT-Tiny can be trimmed down to 50% of the original FLOPs almost without losing accuracy. Codes are available online: https://github.com/VITA-Group/UVC.

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 16 code samples: 13 from 1 repository linked to this paper, 3 identical to code harvested from another paper's repository. Across all 16: 9 ran (of which 3 constructed an object rather than computing a result; 6 with no instrument failure: 0 honoured, 1 violated, 5 with no contract checked; 3 where Syntology's instrument failed) · 7 unverified. Where Syntology's instrument failed: 3 where our contract draft was wrong; failures of Syntology's instrument, not of the code.

By repository: official repository: 13 samples from 1 repository, 7 ran (6 with no instrument failure; 1 where Syntology's instrument failed); 3 identical to code first harvested elsewhere. The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

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. 2 of the 9 samples that ran have no recorded date.

VITA-Group/UVC officialmentioned in paperpytorchSyntology: 13 samples, 7 ran (1 instrument-failed) · 6 unverified 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.

Code Syntology ran Syntology

16 samples harvested: 9 ran (of which 3 constructed an object rather than computing a result; 6 with no instrument failure: 0 honoured, 1 violated, 5 with no contract checked; 3 where Syntology's instrument failed) · 7 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 · violated contract
3ran · our draft was wrong
5ran
7unverified

Licence: 3 of the 16 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 VITA-Group/UVC. 3 samples are listed last, in a group without a repository. “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 3 samples 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.

VITA-Group/UVC official repository 13 samples, 7 ran (1 instrument-failed) · 6 unverified
Attention in 344 papers
census: 13 implementations from 12 papers, 10 distinct outputs across 4 buckets, one shared input per bucket
UVC/models/modeling.py ran · metamorphic tier: deterministic MIT (permissive) · bfc8fe0ddf96d4c7 · report fixture: returned Attention · level recorded 2026-09-22
Mlp in 102 papers UVC/models/modeling.py ran · metamorphic tier: deterministic fingerprinted MIT (permissive) · 3215e1b62b114566 · report fixture: returned Mlp · level recorded 2026-09-22
PreActBottleneck in 6 papers UVC/models/modeling.py ran MIT (permissive) · df180144004badc3 · report fixture: signature x:Tensor/4/float · returned Tensor · level recorded 2026-09-24
ResNetV2 in 5 papers UVC/models/modeling.py ran · metamorphic tier: deterministic MIT (permissive) · e91e85b73a67665a · report fixture: returned ResNetV2 · level recorded 2026-09-22
StdConv2d UVC/models/modeling.py ran fingerprinted MIT (permissive) · eadbbe03b5b32f01 · report fixture: signature x:Tensor/4/float · returned Tensor · level recorded 2026-09-24
get_tau in 5 papers UVC/joint_train.py ran · violated contract fingerprinted MIT (permissive) · aa9253d77220768c · report fixture: signature max_tau:float/0/float, min_tau:float/0/float, ite:int/0/int, total:int/0/int · returned float · level recorded, date not recorded
get_uvc_layers UVC/post_train.py ran · our draft was wrong MIT (permissive) · 80da1ef156cc3a88 · report fixture: signature model:callable/0/?, args:NoneType/0/? · returned tuple · level recorded, date not recorded
Embeddings in 15 papers UVC/models/modeling.py unverified MIT (permissive) · 8020519bcd7e5931 · report
Encoder in 278 papers UVC/models/modeling.py unverified MIT (permissive) · f5a8c54634cf9950 · report
Transformer in 173 papers UVC/models/modeling.py unverified MIT (permissive) · a4485af62ba072e0 · report
VisionTransformer in 53 papers UVC/models/modeling.py unverified MIT (permissive) · 4900a340f86d38d4 · report
complex_accuracy UVC/joint_train.py unverified MIT (permissive) · 3e8918f575824271 · report
Repository not located 3 samples, 2 ran (2 instrument-failed) · 1 unverified
3 samples are identical to code Syntology harvested from another paper's repository; this paper's own copy was not located, so no repository is named and no licence is recorded.
conv1x1 in 691 papers repository not located ran · our draft was wrong licence of this copy not recorded · b0f916df39e78ae5 · report fixture: signature cin:int/0/int, cout:int/0/int, stride:int/0/int, bias:bool/0/bool · returned StdConv2d · level recorded 2026-09-22
conv3x3 in 1,376 papers repository not located ran · our draft was wrong licence of this copy not recorded · 4e5bf13dbdc4f008 · report fixture: signature cin:int/0/int, cout:int/0/int, stride:int/0/int, groups:int/0/int, bias:bool/0/bool · returned StdConv2d · level recorded 2026-09-22
simple_accuracy in 66 papers repository not located unverified licence of this copy not recorded · 3c241ecfe3749a6d · 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 UNIFIED VISUAL TRANSFORMER COMPRESSION (graph method: UNIFIED VISUAL TRANSFORMER COMPRESSION). 14 bricks; individually on a fixture: 9 ran (of which 3 constructed an object rather than computing a result; 5 with no instrument failure: 0 honoured, 0 violated, 5 with no contract checked; 4 where Syntology's instrument failed) · 5 unverified. The assembly has not been run end-to-end.

  1. VisionTransformer entry · level 0 · unverified · 4900a340f86d38d4
  2. Transformer model · level 0 · unverified · a4485af62ba072e0
  3. np2th helper · level 2 · ran; the contract check could not run · 7d8eaf5537f699e4
  4. Embeddings model · level 0 · unverified · 8020519bcd7e5931
  5. Encoder model · level 0 · unverified · f5a8c54634cf9950
  6. ResNetV2 helper · level 1 · ran · e91e85b73a67665a
  7. Block model · level 0 · unverified · 195dcd4e31d9f812
  8. PreActBottleneck helper · level 1 · ran · df180144004badc3
  9. StdConv2d helper · level 1 · ran · eadbbe03b5b32f01
  10. Attention model · level 1 · ran · bfc8fe0ddf96d4c7
  11. Mlp model · level 1 · ran · 3215e1b62b114566
  12. conv1x1 helper · level 1 · ran · our draft was wrong · b0f916df39e78ae5
  13. conv3x3 helper · level 1 · ran · our draft was wrong · 4e5bf13dbdc4f008
  14. swish helper · level 2 · ran · our draft was wrong · 0f786c407fb1ee4c

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="2203.08243") returns these assemblies with each brick's role and wiring (how to connect).

Tasks

Knowledge Distillation

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Methods

Absolute Position EncodingsAdamAttentionAttention DropoutBPEConvolutionDeiTDense ConnectionsDropoutFeedforward NetworkLabel SmoothingLayer NormalizationLinear LayerMulti-Head AttentionPosition-Wise Feed-Forward LayerPruningResidual ConnectionSoftmaxT2T-ViTTransformerVision Transformer

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