SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2204.08446 · ECCV · 2022

VSA: Learning Varied-Size Window Attention in Vision Transformers

Dacheng Tao, Jing Zhang, Yufei Xu, Qiming Zhang

arXiv · PDF · Open in the Atlas

Code that ran

3 of 4 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (3 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 3 have no contract verdict; 0 failures of our instrument; 1 no recorded run), measured 2026-11-22; repository links from the Papers with Code archive attributed under CC BY-SA, no affiliation; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.

"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/2204.08446.json · badge: /api/ran/2204.08446.svg

Each repository's role is a label: a classifier's output recorded on the link, or, where no role is recorded, what the Papers with Code archive marked. The line under it names its basis. Neither says who wrote the code.

RepositoryRoleRan
vitae-transformer/qformer community
Papers with Code archive
2 ran · 0 no recorded run
ViTAE-Transformer/ViTAE-VSA community
Papers with Code archive
1 ran · 0 no recorded run
copy not recorded community
not in the paper's declared repositories
0 ran · 1 no recorded run
FunctionStatusWhere it lives
RectifyCoordsGradient Ran
ran on a synthesized input · constructed an object; no result was computed
vitae-transformer/qformer/models/swin/swin_transformer_vsa.py
code served (permissive licence) · get_code("22ebd1057ed2d0c2")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned an instance of its own class, RectifyCoordsGradient: the run constructed the object and did not call it, so no result was computed. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn

torch.manual_seed(0)

# Create a small tensor for coords parameter
# Shape: (2, 4) - batch of 2, dimension of 4
coords = torch.randn(2, 4, requires_grad=True)

# coords_lambda is a scalar parameter with default value 20
coords_lambda = 20

ARGS = {
    "coords": coords,
    "coords_lambda": coords_lambda
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type RectifyCoordsGradient — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a RectifyCoordsGradient output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
VSAAttention Ran
ran on a synthesized input · constructed an object; no result was computed
vitae-transformer/qformer/models/swin/swin_transformer_vsa.py
code served (permissive licence) · get_code("2f2d2dbf82d3f3db")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned an instance of its own class, VSAAttention: the run constructed the object and did not call it, so no result was computed. The image it ran in was not recorded. The code mentions cuda, and the harness aliases cuda to the CPU for every sample; whether this run needed that is not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn

torch.manual_seed(0)

ARGS = {
    'dim': 64,
    'num_heads': 8,
    'qkv_bias': False,
    'attn_drop': 0.0,
    'proj_drop': 0.0,
    'window_size': 7,
    'rpe': 'v2',
    'coords_lambda': 20,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
einops, timm, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type VSAAttention — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a VSAAttention output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
VSAWindowAttention Ran
ran on a synthesized input · constructed an object; no result was computed
ViTAE-Transformer/ViTAE-VSA/Image-Classification/vitaev2_vsa/vsa.py
pointer only (licence: NONE) · get_code("e01b15632fdf3e8d")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, ViTAE-Transformer/ViTAE-VSA when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

calc_rel_pos_spatial Not yet run
unverified (harvested, no recorded run)
this paper's copy was not recorded; identical code first harvested from vitae-transformer/qformer
pointer only · get_code("477da5846437d6e9")
How it ran

The body of this sample is withheld: this paper's copy of the code was not recorded (its link names no repository or source path), so no licence was read for it. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. No invocation recorded.

Also present in other papers' code

Of 4 samples lifted from this paper's linked repositories, 4 carry a function identity and 0 do not. Of the 4 distinct functions with an identity, 1 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 1 later) and 3 are only here among the 52,977 papers whose code Syntology holds (as of 2026-09-26).

Only here means: among the 52,977 papers whose code Syntology holds (51,887 with lifted samples, 1,090 through their assemblies only), as of 2026-09-26, no other paper's lifted samples or current assembly bricks contain an identical function body after normalisation.

FunctionAlso present inEarliest-datedOthers named
calc_rel_pos_spatial
brick of this paper's assembly: helper
1 other paper
1 later
2303.15105 (2023-03)
Vision Transformer with Quadrangle Attention

Presence, not direction: an identical function body in two papers' repositories does not say who wrote it, which came from which, or that either paper's method rests on it. Matching is exact after normalisation, so a renamed or edited function is not matched. Dates are when papers appeared, not when code was written. Only .py files the harvester lifted are covered; notebooks are not read. A new-style arXiv id YYMM.NNNN dates a paper to the month 20YY-MM; otherwise Paper.year dates it to the year; otherwise it is undated. Dates are when the paper appeared, not when the code was written. Read 2026-11-22.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Attention within windows has been widely explored in vision transformers to balance the performance, computation complexity, and memory footprint. However, current models adopt a hand-crafted fixedsize window design, which restricts their capacity of modeling long-term dependencies and adapting to objects of different sizes. To address this drawback, we propose Varied-Size Window Attention (VSA) to learn adaptive window configurations from data. Specifically, based on the tokens within each default window, VSA employs a window regression module to predict the size and location of the target window, i.e., the attention area where the key and value tokens are sampled. By adopting VSA independently for each attention head, it can model long-term dependencies, capture rich context from diverse windows, and promote information exchange among overlapped windows. VSA is an easy-toimplement module that can replace the window attention in state-ofthe-art representative models with minor modifications and negligible extra computational cost while improving their performance by a large margin, e.g., 1.1% for Swin-T on ImageNet classification. In addition, the performance gain increases when using larger images for training and test. Experimental results on more downstream tasks, including object detection, instance segmentation, and semantic segmentation, further demonstrate the superiority of VSA over the vanilla window attention in dealing with objects of different sizes. The code is available at https://github.com/ViTAE-Transformer/ViTAE-VSA.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2204.08446")
get_code_for_paper("2204.08446")
have("2204.08446")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/2204.08446.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/2204.08446.svg)](https://syntology.ai/paper/2204.08446)

Connect an agent — have() is free.