SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2106.13112 · 2021

VOLO: Vision Outlooker for Visual Recognition

arXiv · PDF · Open in the Atlas

Code that ran

5 of 6 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (0 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 0 have no contract verdict; 5 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/2106.13112.json · badge: /api/ran/2106.13112.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
sail-sg/volo canonical (classifier label)
classifier basis: census
5 ran · 1 no recorded run
FunctionStatusWhere it lives
get_block Ran
ran; our contract draft was wrong, not the code
sail-sg/volo/models/volo.py
code served (permissive licence) · get_code("87a34d0211098d18")
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 ClassBlock. 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
from timm.models.layers import DropPath, to_2tuple, trunc_normal_

torch.manual_seed(0)

ARGS = {
    'block_type': 'ca',
    'dim': 64,
    'num_heads': 4,
    'head_dim': 16,
    'mlp_ratio': 4.0,
    'qkv_bias': False,
    'qk_scale': None,
    'drop': 0.1,
    'attn_drop': 0.1,
    'drop_path': 0.1,
    'act_layer': nn.GELU,
    'norm_layer': nn.LayerNorm,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
timm, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ClassBlock — a ClassBlock output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a ClassBlock 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
outlooker_blocks Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
sail-sg/volo/models/volo.py
code served (permissive licence) · get_code("ba7779c554538760")
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 Sequential. 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)

# Define a minimal block_fn that the function will call
class MinimalOutlookerBlock(nn.Module):
    def __init__(self, dim, kernel_size=3, padding=1, stride=1, num_heads=1, mlp_ratio=3.0, qkv_bias=False, qk_scale=None, attn_drop=0, drop_path=0.0):
        super().__init__()
        self.dim = dim
        self.kernel_size = kernel_size
        self.padding = padding
        self.stride = stride
        self.num_heads = num_heads
        self.mlp_ratio = mlp_ratio
        self.qkv_bias = qkv_bias
        self.qk_scale = qk_scale
        self.attn_drop = attn_drop
        self.drop_path = drop_path
        self.linear = nn.Linear(dim, dim)
    
    def forward(self, x):
        return self.linear(x)

# Define ARGS dictionary with all parameters
ARGS = {
    'block_fn': MinimalOutlookerBlock,
    'index': 0,
    'dim': 64,
    'layers': [2, 3, 4, 2],
    'num_heads': 4,
    'kernel_size': 3,
    'padding': 1,
    'stride': 1,
    'mlp_ratio': 3.0,
    'qkv_bias': False,
    'qk_scale': None,
    'attn_drop': 0.0,
    'drop_path_rate': 0.1,
}
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 Sequential — a Sequential output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Sequential output the recorded ndim 1 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
resize_pos_embed Ran
ran; our contract draft was wrong, not the code
sail-sg/volo/utils/utils.py
code served (permissive licence) · get_code("b3b01c5690550b6c")
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 Tensor, ndim 3, dtype family float, shape (1, 577, 4) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape 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 math

torch.manual_seed(0)

# posemb: original position embedding tensor with shape (batch, num_tokens, embedding_dim)
# For example: (1, 1 + 14*14, 768) where 1 is class token and 14x14 is grid
posemb = torch.randn(1, 197, 4)  # 1 + 196 (14x14) tokens, embedding_dim=4

# posemb_new: target position embedding tensor with shape (batch, num_tokens_new, embedding_dim)
# For example: (1, 1 + 24*24, 768) where 1 is class token and 24x24 is grid
posemb_new = torch.randn(1, 577, 4)  # 1 + 576 (24x24) tokens, embedding_dim=4

ARGS = {
    "posemb": posemb,
    "posemb_new": posemb_new,
}
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 Tensor, ndim 3, dtype family float, shape (1, 577, 4) — shape (1, 577, 4) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
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
resize_pos_embed_4d Ran
ran; our contract draft was wrong, not the code
sail-sg/volo/utils/utils.py
code served (permissive licence) · get_code("e0d164428f4b2b84")
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 Tensor, ndim 4, dtype family float, shape (2, 3, 3, 5) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape 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.functional as F

torch.manual_seed(0)

# Create position embedding tensors with shape (batch, grid_size, grid_size, channels)
# Using distinct dimensions: batch=2, old_grid=4, new_grid=3, channels=5
posemb = torch.randn(2, 4, 4, 5)
posemb_new = torch.randn(2, 3, 3, 5)

ARGS = {
    'posemb': posemb,
    'posemb_new': posemb_new,
}
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 Tensor, ndim 4, dtype family float, shape (2, 3, 3, 5) — shape (2, 3, 3, 5) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
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
resize_pos_embed_without_cls Ran
ran; our contract draft was wrong, not the code
sail-sg/volo/utils/utils.py
code served (permissive licence) · get_code("6aee19396cb78860")
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 Tensor, ndim 3, dtype family float, shape (1, 576, 768) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape 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 math

torch.manual_seed(0)

# Create position embeddings for a smaller grid (14x14 = 196 tokens)
# Shape: (1, 196, 768) - batch_size=1, num_tokens=196, embedding_dim=768
gs_old = 14
posemb = torch.randn(1, gs_old * gs_old, 768)

# Create new position embedding target with larger grid (24x24 = 576 tokens)
# Shape: (1, 576, 768) - batch_size=1, num_tokens=576, embedding_dim=768
gs_new = 24
posemb_new = torch.randn(1, gs_new * gs_new, 768)

ARGS = {
    'posemb': posemb,
    'posemb_new': posemb_new,
}
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 Tensor, ndim 3, dtype family float, shape (1, 576, 768) — shape (1, 576, 768) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
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
rand_bbox Not yet run
unverified (harvested, no recorded run)
sail-sg/volo/models/volo.py
code served (permissive licence) · get_code("b66088e90761117d")
How it ran

no invocation recorded.

Also present in other papers' code

Of 6 samples lifted from this paper's linked repositories, 6 carry a function identity and 0 do not. Of the 6 distinct functions with an identity, 6 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 6 later) and 0 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
rand_bbox 3 other papers
3 later
2203.14509 (2022-03)
Automated Progressive Learning for Efficient Training of Vision Transformers
2204.00993 (2022-04)
Improving Vision Transformers by Revisiting High-frequency Components
2207.04978 (2022-07)
Wave-ViT: Unifying Wavelet and Transformers for Visual Representation Learning
get_block 2 other papers
2 later
2203.14509 (2022-03)
Automated Progressive Learning for Efficient Training of Vision Transformers
2204.00993 (2022-04)
Improving Vision Transformers by Revisiting High-frequency Components
outlooker_blocks 2 other papers
2 later
2203.14509 (2022-03)
Automated Progressive Learning for Efficient Training of Vision Transformers
2204.00993 (2022-04)
Improving Vision Transformers by Revisiting High-frequency Components
resize_pos_embed 1 other paper
1 later
2206.07662 (2022-06)
SP-ViT: Learning 2D Spatial Priors for Vision Transformers
resize_pos_embed_4d 1 other paper
1 later
2206.07662 (2022-06)
SP-ViT: Learning 2D Spatial Priors for Vision Transformers
resize_pos_embed_without_cls 1 other paper
1 later
2206.07662 (2022-06)
SP-ViT: Learning 2D Spatial Priors for Vision Transformers

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

Visual recognition has been dominated by convolutional neural networks (CNNs) for years. Though recently the prevailing vision transformers (ViTs) have shown great potential of self-attention based models in ImageNet classification, their performance is still inferior to that of the latest SOTA CNNs if no extra data are provided. In this work, we try to close the performance gap and demonstrate that attention-based models are indeed able to outperform CNNs. We find a major factor limiting the performance of ViTs for ImageNet classification is their low efficacy in encoding fine-level features into the token representations. To resolve this, we introduce a novel outlook attention and present a simple and general architecture, termed Vision Outlooker (VOLO). Unlike self-attention that focuses on global dependency modeling at a coarse level, the outlook attention efficiently encodes finer-level features and contexts into tokens, which is shown to be critically beneficial to recognition performance but largely ignored by the self-attention. Experiments show that our VOLO achieves 87.1% top-1 accuracy on ImageNet-1K classification, which is the first model exceeding 87% accuracy on this competitive benchmark, without using any extra training data In addition, the pre-trained VOLO transfers well to downstream tasks, such as semantic segmentation. We achieve 84.3% mIoU score on the cityscapes validation set and 54.3% on the ADE20K validation set. Code is available at \url{https://github.com/sail-sg/volo}.

For agents

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

get_harvested_code_for_paper("2106.13112")
get_code_for_paper("2106.13112")
have("2106.13112")

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

curl https://syntology.ai/api/ran/2106.13112.json

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

[![Syntology run record](https://syntology.ai/api/ran/2106.13112.svg)](https://syntology.ai/paper/2106.13112)

Connect an agent — have() is free.