SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2106.11539 · ICCV · 2021

DocFormer: End-to-End Transformer for Document Understanding

Yusheng Xie, Srikar Appalaraju, Bhavan Jasani, Bhargava Urala Kota, R. Manmatha

arXiv · PDF · Open in the Atlas

Code that ran

5 of 10 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (5 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 5 have no contract verdict; 0 failures of our instrument; 5 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/2106.11539.json · badge: /api/ran/2106.11539.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
shabie/docformer community
Papers with Code archive
5 ran · 5 no recorded run
FunctionStatusWhere it lives
MultiModalAttentionLayer Ran
ran on a synthesized input · constructed an object; no result was computed
shabie/docformer/src/docformer/modeling.py
code served (permissive licence) · get_code("08d9b843959a11da")
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, MultiModalAttentionLayer: 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)

# Parameters for MultiModalAttentionLayer
embed_dim = 8
n_heads = 2
max_relative_position = 4
max_seq_length = 5
dropout = 0.1

ARGS = {
    "embed_dim": embed_dim,
    "n_heads": n_heads,
    "max_relative_position": max_relative_position,
    "max_seq_length": max_seq_length,
    "dropout": dropout,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
einops, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type MultiModalAttentionLayer — 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 MultiModalAttentionLayer 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
PositionalEncoding Ran
ran on a synthesized input · constructed an object; no result was computed
shabie/docformer/src/docformer/modeling.py
code served (permissive licence) · get_code("95c11b9646ef2995")
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, PositionalEncoding: 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 math
import torch
import torch.nn as nn

torch.manual_seed(0)

ARGS = {
    "d_model": 4,
    "dropout": 0.1,
    "max_len": 5000,
}
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 PositionalEncoding — 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 PositionalEncoding 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
PreNormAttn Ran
ran on a synthesized input · constructed an object; no result was computed
shabie/docformer/src/docformer/modeling.py
code served (permissive licence) · get_code("9d44110b605b91c8")
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, PreNormAttn: 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 simple attention function that takes 4 normalized inputs
class SimpleAttention(nn.Module):
    def __init__(self):
        super().__init__()
    
    def forward(self, norm_t_bar, norm_v_bar, norm_t_bar_s, norm_v_bar_s, **kwargs):
        # Simple attention: just return a combination of inputs
        return norm_t_bar + norm_v_bar + norm_t_bar_s + norm_v_bar_s

# Set up dimensions
dim = 4
batch_size = 2

# Create the attention function
attention_fn = SimpleAttention()

ARGS = {
    "dim": dim,
    "fn": attention_fn,
}
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 PreNormAttn — 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 PreNormAttn 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
RelativePosition Ran
ran on a synthesized input · constructed an object; no result was computed
shabie/docformer/src/docformer/modeling.py
code served (permissive licence) · get_code("2485d54f61db9f8d")
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, RelativePosition: 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)

ARGS = {
    "num_units": 4,
    "max_relative_position": 3,
    "max_seq_length": 8,
}
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 RelativePosition — 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 RelativePosition 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
ResNetFeatureExtractor Ran
ran on a synthesized input · constructed an object; no result was computed
shabie/docformer/src/docformer/modeling.py
code served (permissive licence) · get_code("3c51327ee452d265")
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, ResNetFeatureExtractor: 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)

ARGS = {
    'hidden_dim': 256
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
einops, torch, torchvision — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ResNetFeatureExtractor — 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 ResNetFeatureExtractor 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
DocFormer Not yet run
unverified (harvested, no recorded run)
shabie/docformer/src/docformer/modeling.py
code served (permissive licence) · get_code("b7af5dd1c0383ff1")
How it ran

no invocation recorded.

DocFormerEmbeddings Not yet run
unverified (harvested, no recorded run)
shabie/docformer/src/docformer/modeling.py
code served (permissive licence) · get_code("4006b85007037117")
How it ran

no invocation recorded.

DocFormerEncoder Not yet run
unverified (harvested, no recorded run)
shabie/docformer/src/docformer/modeling.py
code served (permissive licence) · get_code("682dbc1c67b2b9df")
How it ran

no invocation recorded.

ExtractFeatures Not yet run
unverified (harvested, no recorded run)
shabie/docformer/src/docformer/modeling.py
code served (permissive licence) · get_code("714cd85a4f672596")
How it ran

no invocation recorded.

LanguageFeatureExtractor Not yet run
unverified (harvested, no recorded run)
shabie/docformer/src/docformer/modeling.py
code served (permissive licence) · get_code("330baaddd559fac9")
How it ran

no invocation recorded.

Also present in other papers' code

Of 10 samples lifted from this paper's linked repositories, 10 carry a function identity and 0 do not. Of the 10 distinct functions with an identity, 0 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 0 later) and 10 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.

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

We present DocFormer -a multi-modal transformer based architecture for the task of Visual Document Understanding (VDU). VDU is a challenging problem which aims to understand documents in their varied formats (forms, receipts etc.) and layouts. In addition, DocFormer is pre-trained in an unsupervised fashion using carefully designed tasks which encourage multi-modal interaction. DocFormer uses text, vision and spatial features and combines them using a novel multi-modal self-attention layer. DocFormer also shares learned spatial embeddings across modalities which makes it easy for the model to correlate text to visual tokens and vice versa. DocFormer is evaluated on 4 different datasets each with strong baselines. DocFormer achieves state-of-the-art results on all of them, sometimes beating models 4x its size (in no. of parameters).

For agents

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

get_harvested_code_for_paper("2106.11539")
get_code_for_paper("2106.11539")
have("2106.11539")

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

curl https://syntology.ai/api/ran/2106.11539.json

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

[![Syntology run record](https://syntology.ai/api/ran/2106.11539.svg)](https://syntology.ai/paper/2106.11539)

Connect an agent — have() is free.