Papers › SegFormer: Simple and Efficient Design for Semantic Segmentation with Transformers

SegFormer: Simple and Efficient Design for Semantic Segmentation with Transformers

31 May 2021NeurIPS 2021 12arXiv:2105.15203archive 2025-07-28

Enze Xie, Wenhai Wang, Zhiding Yu, Anima Anandkumar, Jose M. Alvarez, Ping Luo

We present SegFormer, a simple, efficient yet powerful semantic segmentation framework which unifies Transformers with lightweight multilayer perception (MLP) decoders. SegFormer has two appealing features: 1) SegFormer comprises a novel hierarchically structured Transformer encoder which outputs multiscale features. It does not need positional encoding, thereby avoiding the interpolation of positional codes which leads to decreased performance when the testing resolution differs from training. 2) SegFormer avoids complex decoders. The proposed MLP decoder aggregates information from different layers, and thus combining both local attention and global attention to render powerful representations. We show that this simple and lightweight design is the key to efficient segmentation on Transformers. We scale our approach up to obtain a series of models from SegFormer-B0 to SegFormer-B5, reaching significantly better performance and efficiency than previous counterparts. For example, SegFormer-B4 achieves 50.3% mIoU on ADE20K with 64M parameters, being 5x smaller and 2.2% better than the previous best method. Our best model, SegFormer-B5, achieves 84.0% mIoU on Cityscapes validation set and shows excellent zero-shot robustness on Cityscapes-C. Code will be released at: github.com/NVlabs/SegFormer.

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

The archive's official repository (NVlabs/SegFormer) was not harvested; the 85 samples below come from 11 community repositories, and 1 more is identical to code Syntology harvested from another paper's repository.

Syntology 86 code samples: 85 from 11 repositories linked to this paper, 1 identical to code harvested from another paper's repository. Across all 86: 66 ran (of which 22 constructed an object rather than computing a result; 52 with no instrument failure: 1 honoured, 2 violated, 49 with no contract checked; 14 where Syntology's instrument failed) · 20 unverified. Where Syntology's instrument failed: 12 where our contract draft was wrong; 2 where the contract check could not run; failures of Syntology's instrument, not of the code.

By repository: community (archive-listed): 83 samples from 10 repositories, 63 ran (51 with no instrument failure; 12 where Syntology's instrument failed); community: 2 samples from 1 repository, 2 ran (0 with no instrument failure; 2 where Syntology's instrument failed); 1 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.

No sample is linked to this paper from 5 of its repositories that Syntology harvested for other papers:

Levels written between 2026-09-22 and 2026-09-25 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed. 2 of the 66 samples that ran have no recorded date.

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

NVlabs/SegFormer officialmentioned in papermentioned on GitHubpytorchNOASSERTIONSyntology: not harvested report
DarshanDeshpande/jax-models mentioned on GitHubjaxSyntology: 9 samples, 5 ran (0 instrument-failed) · 4 unverified report
GhadeerElmkaiel/SegFormer mentioned on GitHubpytorchNOASSERTIONSyntology: not harvested report
IMvision12/SegFormer-tf mentioned on GitHubtfSyntology: 12 samples, 9 ran (0 instrument-failed) · 3 unverified report
IMvision12/keras-vision-models mentioned on GitHubpytorchSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Swin Transformer: Hierarchical Vision Transformer using Shifted Windows (arXiv:2103.14030; that paper's own run record: 123 ran (of which 45 constructed an object rather than computing a result; 79 with no instrument failure: 6 honoured, 3 violated, 70 with no contract checked; 44 where Syntology's instrument failed) · 84 unverified); nothing checks that they implement this paper's method. report
UAws/CV-3315-Is-All-You-Need mentioned on GitHubpytorchApache-2.0Syntology: 7 samples, 6 ran (3 instrument-failed) · 1 unverified report
VikParuchuri/surya mentioned on GitHubpytorchApache-2.0Syntology: 9 samples, 8 ran (5 instrument-failed) · 1 unverified report
alessiomora/unlearning_fl mentioned on GitHubtfApache-2.0Syntology: 13 samples, 10 ran (0 instrument-failed) · 3 unverified report
chenghan111/dnc mentioned on GitHubpytorchSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Visual Recognition with Deep Nearest Centroids (arXiv:2209.07383; that paper's own run record: 3 ran (of which 0 constructed an object rather than computing a result; 0 with no instrument failure: 0 honoured, 0 violated, 0 with no contract checked; 3 where Syntology's instrument failed) · 0 unverified); nothing checks that they implement this paper's method. report
dzhaoxd/seco mentioned on GitHubpytorchSyntology: not harvested report
huggingface/transformers mentioned on GitHubpytorchSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Attention Is All You Need (arXiv:1706.03762; that paper's own run record: 610 ran (of which 293 constructed an object rather than computing a result; 529 with no instrument failure: 45 honoured, 15 violated, 469 with no contract checked; 81 where Syntology's instrument failed) · 336 unverified), HuggingFace's Transformers: State-of-the-art Natural Language Processing (arXiv:1910.03771; that paper's own run record: 3 ran (of which 0 constructed an object rather than computing a result; 0 with no instrument failure: 0 honoured, 0 violated, 0 with no contract checked; 3 where Syntology's instrument failed) · 0 unverified), DeBERTa: Decoding-enhanced BERT with Disentangled Attention (arXiv:2006.03654; that paper's own run record: 4 ran (of which 0 constructed an object rather than computing a result; 3 with no instrument failure: 2 honoured, 1 violated, 0 with no contract checked; 1 where Syntology's instrument failed) · 9 unverified) and 1 more (every one in this page's JSON twin); nothing checks that they implement this paper's method. report
kikacaty/RAP_Benchmark mentioned on GitHubpytorchNOASSERTIONSyntology: not harvested report
lucidrains/segformer-pytorch mentioned on GitHubpytorchSyntology: 8 samples, 6 ran (0 instrument-failed) · 2 unverified report
macdonaldezra/minesegsat mentioned on GitHubpytorchSyntology: 7 samples, 5 ran (0 instrument-failed) · 2 unverified report
sithu31296/semantic-segmentation mentioned on GitHubpytorchSyntology: 5 samples, 4 ran (0 instrument-failed) · 1 unverified report
toqitahamid/gasformer mentioned on GitHubpytorchSyntology: not harvested report
2023-MindSpore-4/Code11 mindsporeSyntology: not harvested report
2023-MindSpore-4/Code11 mindsporeSyntology: not harvested report
2023-MindSpore-4/Code14 mindsporeSyntology: not harvested report
BR-IDL/PaddleViT paddleSyntology: not harvested report
DavidLandup0/deepvision pytorchSyntology: 7 samples, 4 ran (2 instrument-failed) · 3 unverified report
JoegameZhou/Segmenter_workers mindsporeApache-2.0Syntology: 6 samples, 6 ran (2 instrument-failed) · 0 unverified report
KieDani/SegformerPlusPlus pytorchGPL-3.0Syntology: not harvested report
MindSpore-paper-code-3/code5 mindsporeSyntology: not harvested report
MindSpore-paper-code-3/code5 mindsporeSyntology: not harvested report
PaddlePaddle/PaddleSeg paddleSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to EdgeFlow: Achieving Practical Interactive Segmentation with Edge-Guided Flow (arXiv:2109.09406; that paper's own run record: 5 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; 2 where Syntology's instrument failed) · 0 unverified); nothing checks that they implement this paper's method. report
code-implementation1/Code8 mindsporeSyntology: not harvested report
pwc-1/Paper-10 mindsporeSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to StableToolBench: Towards Stable Large-Scale Benchmarking on Tool Learning of Large... (arXiv:2403.07714; that paper's own run record: 11 ran (of which 0 constructed an object rather than computing a result; 8 with no instrument failure: 0 honoured, 2 violated, 6 with no contract checked; 3 where Syntology's instrument failed) · 4 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

86 samples harvested: 66 ran (of which 22 constructed an object rather than computing a result; 52 with no instrument failure: 1 honoured, 2 violated, 49 with no contract checked; 14 where Syntology's instrument failed) · 20 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page last changed ; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

1ran · honoured contract
2ran · violated contract
12ran · our draft was wrong
2ran; the contract check could not run
49ran
20unverified

Licence: 20 of the 86 samples are pointer only, meaning Syntology does not serve that copy's text. For 5 of them, this paper's copy is under a permissive licence and cleared, but the stored copy of the same code that Syntology serves is not; the licence label's tooltip gives Syntology's recorded reason. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from 11 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. 1 sample is 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 4 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.

DarshanDeshpande/jax-models community (archive-listed) 9 samples, 5 ran (0 instrument-failed) · 4 unverified
DepthwiseConv2D in 8 papers jax_models/models/segformer.py ran Apache-2.0 (permissive) · ccadeb436e01c85c · report fixture: returned DepthwiseConv2D · level recorded 2026-09-22
DropPath in 75 papers
census: 17 implementations from 17 papers, 8 distinct outputs across 3 buckets, one shared input per bucket
jax_models/models/segformer.py ran Apache-2.0 (permissive) · 8085cbc641b30993 · report fixture: returned DropPath · level recorded 2026-09-22
MLPHead in 5 papers jax_models/models/segformer.py ran Apache-2.0 (permissive) · 1cca7df806bbd21f · report fixture: returned MLPHead · level recorded 2026-09-22
SegFormerHead jax_models/models/segformer.py ran Apache-2.0 (permissive) · 6caaf2219052f191 · report fixture: returned SegFormerHead · level recorded 2026-09-22
TransformerMLP in 5 papers jax_models/models/segformer.py ran Apache-2.0 (permissive) · 6afe61dba51eef77 · report fixture: returned TransformerMLP · level recorded 2026-09-22
EfficientSelfAttention jax_models/models/segformer.py unverified Apache-2.0 (permissive) · 1bfaa1f3894b09c5 · report
MixTransformer jax_models/models/segformer.py unverified Apache-2.0 (permissive) · e5084c52a530491b · report
SegFormer jax_models/models/segformer.py unverified Apache-2.0 (permissive) · d2ba4808c201185e · report
IMvision12/SegFormer-tf community (archive-listed) 12 samples, 9 ran (0 instrument-failed) · 3 unverified
Attention in 344 papers
census: 13 implementations from 12 papers, 10 distinct outputs across 4 buckets, one shared input per bucket
models/segformer.py ran no licence file found · pointer only · c26b3dcb83805d8b · report fixture: returned Attention · level recorded 2026-09-22
Block in 252 papers
census: 5 implementations from 5 papers, 3 distinct outputs across 2 buckets, one shared input per bucket
models/segformer.py ran no licence file found · pointer only · 0730c486d30ba0ad · report fixture: returned Block · level recorded 2026-09-22
ConvModule in 8 papers models/segformer.py ran no licence file found · pointer only · 985e0029300d8b09 · report fixture: returned ConvModule · level recorded 2026-09-22
DWConv in 14 papers models/segformer.py ran no licence file found · pointer only · 4817e9583fbf6b3c · report fixture: returned DWConv · level recorded 2026-09-22
DropPath in 75 papers
census: 17 implementations from 17 papers, 8 distinct outputs across 3 buckets, one shared input per bucket
models/segformer.py ran fingerprinted no licence file found · pointer only · 4b04a139b83d832e · report fixture: signature x:EagerTensor/3/float, training:bool/0/bool · returned EagerTensor · level recorded 2026-09-24
MLP in 375 papers models/segformer.py ran no licence file found · pointer only · 7ff5661a4c26fcb7 · report fixture: returned MLP · level recorded 2026-09-22
Mlp in 102 papers models/segformer.py ran no licence file found · pointer only · 635ca5ddeb4973f7 · report fixture: returned Mlp · level recorded 2026-09-22
OverlapPatchEmbed in 16 papers models/segformer.py ran no licence file found · pointer only · 31c432acc4aa0d25 · report fixture: returned OverlapPatchEmbed · level recorded 2026-09-22
SegFormerHead models/segformer.py ran no licence file found · pointer only · bc90afe01ade811b · report fixture: returned SegFormerHead · level recorded 2026-09-22
MixVisionTransformer models/segformer.py unverified no licence file found · pointer only · d47ffdc9c58366e1 · report
ResizeLayer models/segformer.py unverified no licence file found · pointer only · 51b3f94442e3e993 · report
SegFormer_B0 models/segformer.py unverified no licence file found · pointer only · 41a9c09a3f658a00 · report
UAws/CV-3315-Is-All-You-Need community (archive-listed) 7 samples, 6 ran (3 instrument-failed) · 1 unverified
calculate_file_sha256 in 10 papers .dev/gather_models.py ran · our draft was wrong Apache-2.0 (permissive) here; the stored copy is pointer only · a669df2cd07c07bb · report fixture: signature file_path:str/0/str · returned str · level recorded 2026-09-24
execute_now bridge/executor.py ran fingerprinted Apache-2.0 (permissive) · 55b8c7ef424d2298 · report fixture: signature command:str/0/str · returned tuple · level recorded 2026-09-24
get_final_iter in 15 papers .dev/gather_models.py ran · honoured contract Apache-2.0 (permissive) here; the stored copy is pointer only · 88d68650d669bb11 · report fixture: signature config:str/0/str · returned int · level recorded 2026-09-24
process_checkpoint in 87 papers .dev/gather_models.py ran · our draft was wrong Apache-2.0 (permissive) · aa74cd62302ed7c8 · report fixture: signature in_file:str/0/str, out_file:str/0/str · returned str · level recorded 2026-09-24
process_model_info in 21 papers .dev/generate_benchmark_evaluation_script.py ran · our draft was wrong Apache-2.0 (permissive) · 39403bf2697d1720 · report fixture: signature model_info:dict/0/?, work_dir:str/0/str · returned dict · level recorded 2026-09-24
run_command in 61 papers bridge/executor.py ran fingerprinted Apache-2.0 (permissive) · 744dc51a3ef15686 · report fixture: signature command:str/0/str · returned tuple · level recorded 2026-09-24
check_url in 17 papers .dev/check_urls.py unverified Apache-2.0 (permissive) here; the stored copy is pointer only · dbbc38f85ae9361f · report
VikParuchuri/surya community (archive-listed) 9 samples, 8 ran (5 instrument-failed) · 1 unverified
expand_bbox in 11 papers surya/common/util.py ran · violated contract fingerprinted Apache-2.0 (permissive) · 47e89f4fa09782a6 · report fixture: signature bbox:list/1/float, expansion_factor:float/0/float · returned list · level recorded 2026-09-24
find_pruneable_heads_and_indices in 19 papers surya/common/compat.py ran Apache-2.0 (permissive) · cb0ea0c51176639b · report fixture: signature heads:list/1/int, n_heads:int/0/int, head_size:int/0/int, already_pruned_heads:set/0/? · returned tuple · level recorded 2026-09-24
get_aligned_output_features_output_indices surya/common/compat.py ran Apache-2.0 (permissive) · e03be3562fa16493 · report fixture: signature out_features:NoneType/0/?, out_indices:NoneType/0/?, stage_names:list/1/str · returned tuple · level recorded 2026-09-24
get_model_name in 34 papers surya/common/s3.py ran · our draft was wrong fingerprinted Apache-2.0 (permissive) · 072e0940fa122b3f · report fixture: signature pretrained_model_name_or_path:str/0/str · returned str · level recorded 2026-09-24
is_blank_region surya/common/blank.py ran · our draft was wrong Apache-2.0 (permissive) · 049c52930f551191 · report fixture: signature image:Image/0/?, white_threshold:int/0/int, blank_pixel_fraction:float/0/float, uniform_color_std:float/0/float · returned bool · level recorded 2026-09-24
join_urls surya/common/s3.py ran · our draft was wrong fingerprinted Apache-2.0 (permissive) · e27103cbb1fca337 · report fixture: signature url1:str/0/str, url2:str/0/str · returned str · level recorded 2026-09-24
near_white_fraction surya/common/blank.py ran · our draft was wrong Apache-2.0 (permissive) · 8acf036b36a65cd2 · report fixture: signature image:Image/0/?, white_threshold:int/0/int · returned float · level recorded 2026-09-24
resolve_model_dir surya/common/rfdetr_torch.py ran; the contract check could not run (DEP_MISSING) Apache-2.0 (permissive) · de9345b90f9f1c07 · report fixture: signature checkpoint:str/0/str · returned str · level recorded 2026-09-24
load_detector surya/common/rfdetr_torch.py unverified Apache-2.0 (permissive) · 95daffa5faab5036 · report
alessiomora/unlearning_fl community (archive-listed) 13 samples, 10 ran (0 instrument-failed) · 3 unverified
attach_classifier_head unlearning_fl/model_utility.py ran Apache-2.0 (permissive) · 384604872a30dccd · report fixture: signature inputs:KerasTensor/0/float, outputs:KerasTensor/0/float, num_classes:int/0/int, classifier_hidden_layers:int/0/int, classifier_unit_pl:int/0/int, random_seed:int/0/int · returned Functional · level recorded 2026-09-24
create_resnet18 unlearning_fl/models.py ran Apache-2.0 (permissive) · 3756f8840c9ca583 · report fixture: signature num_classes:int/0/int, input_shape:tuple/1/NoneType, norm:str/0/str, l2_weight_decay:float/0/float, seed:int/0/int · returned ResNet18 · level recorded 2026-09-24
get_norm_layer in 111 papers unlearning_fl/models.py ran Apache-2.0 (permissive) · 4581a201b7093062 · report fixture: signature norm:str/0/str, channel_axis:int/0/int · returned BatchNormalization · level recorded 2026-09-24
get_normalization_fn unlearning_fl/dataset.py ran Apache-2.0 (permissive) · 247223d98d751eff · report fixture: signature model_name:str/0/str, dataset:str/0/str · returned function · level recorded 2026-09-24
load_client_datasets_from_files unlearning_fl/dataset.py ran Apache-2.0 (permissive) · 7259ad6db5a66899 · report fixture: signature dataset:str/0/str, sampled_client:int/0/int, batch_size:int/0/int, total_clients:int/0/int, alpha:float/0/float, split:str/0/str, model_name:str/0/str, is_training:bool/0/bool, seed:int/0/int · returned _PrefetchDataset · level recorded 2026-09-24
pil_loader in 144 papers unlearning_fl/centralized_aircrafts.py ran Apache-2.0 (permissive) · d2ab609def2d3e40 · report fixture: signature path:str/0/str · returned Image · level recorded 2026-09-24
pil_loader in 144 papers unlearning_fl/dataset_preparation.py ran Apache-2.0 (permissive) · 4fb2c2bdfa4534bd · report fixture: signature path:str/0/str · returned Image · level recorded 2026-09-24
read_fedmlb_distribution unlearning_fl/dataset_preparation.py ran Apache-2.0 (permissive) · 5cb938e6605fcde4 · report fixture: signature dataset:str/0/str, total_clients:int/0/int, alpha:float/0/float, data_folder:str/0/str · returned list · level recorded 2026-09-24
remove_list_from_list unlearning_fl/create_dirichlet_distrib_files.py ran fingerprinted Apache-2.0 (permissive) · 11277e8612b345a3 · report fixture: signature orig_list:list/1/int, to_remove:list/1/int · returned list · level recorded 2026-09-24
set_non_trainable_blocks unlearning_fl/model_utility.py ran Apache-2.0 (permissive) · b9e87a3f669811b6 · report fixture: signature model:callable/0/?, non_trainable_blocks:int/0/int · returned MockModel · level recorded 2026-09-24
create_resnet18_mlb unlearning_fl/models.py unverified Apache-2.0 (permissive) · 1301e7bceb1a009d · report
download_and_preprocess unlearning_fl/dataset_preparation.py unverified Apache-2.0 (permissive) · 2344612e05e368a4 · report
load_selected_client_statistics unlearning_fl/dataset.py unverified Apache-2.0 (permissive) · ac368bd91abe1f93 · report
lucidrains/segformer-pytorch community (archive-listed) 8 samples, 6 ran (0 instrument-failed) · 2 unverified
DsConv2d segformer_pytorch/segformer_pytorch.py ran fingerprinted MIT (permissive) · 10e1daaea8547fb9 · report fixture: returned DsConv2d · level recorded 2026-09-22
EfficientSelfAttention segformer_pytorch/segformer_pytorch.py ran MIT (permissive) · 542396928de7e36a · report fixture: signature x:Tensor/4/float · returned Tensor · level recorded 2026-09-24
MiT segformer_pytorch/segformer_pytorch.py ran MIT (permissive) · a3c01f13dc01b670 · report fixture: returned MiT · level recorded 2026-09-22
MixFeedForward segformer_pytorch/segformer_pytorch.py ran MIT (permissive) · ff4bb124dc115d63 · report fixture: returned MixFeedForward · level recorded 2026-09-22
PreNorm in 34 papers segformer_pytorch/segformer_pytorch.py ran MIT (permissive) · 748000c608a20c77 · report fixture: returned PreNorm · level recorded 2026-09-22
Segformer segformer_pytorch/segformer_pytorch.py ran MIT (permissive) · ce46b8418261110a · report fixture: returned Segformer · level recorded 2026-09-22
cast_tuple in 58 papers segformer_pytorch/segformer_pytorch.py unverified MIT (permissive) · e33cebee6724c045 · report
macdonaldezra/minesegsat community (archive-listed) 7 samples, 5 ran (0 instrument-failed) · 2 unverified
EfficientSelfAttention mine_seg_sat/models/segformer.py ran fingerprinted MIT (permissive) · 5049266ef99b4d26 · report fixture: signature hidden_states:Tensor/3/float, height:int/0/int, width:int/0/int, return_attn:bool/0/bool · returned Tensor · level recorded 2026-09-24
MixedFeedForwardNetwork mine_seg_sat/models/segformer.py ran MIT (permissive) · 46464f58f54aa5a0 · report fixture: signature x:Tensor/3/float, height:int/0/int, width:int/0/int · returned Tensor · level recorded 2026-09-24
OverlapPatchEmbedding mine_seg_sat/models/segformer.py ran · metamorphic tier: deterministic fingerprinted MIT (permissive) · 79ded40dd7812caf · report fixture: signature x:Tensor/4/float · returned tuple · level recorded 2026-09-22
SegFormerDecoder mine_seg_sat/models/segformer.py ran · metamorphic tier: well formed MIT (permissive) · c1cec60a6c890710 · report fixture: signature x:list/1/Tensor · returned Tensor · level recorded 2026-09-22
TransformerBlock in 73 papers mine_seg_sat/models/segformer.py ran MIT (permissive) · 89979e961601b054 · report fixture: signature x:Tensor/3/float, height:int/0/int, width:int/0/int, return_attn:bool/0/bool · returned Tensor · level recorded 2026-09-24
SegFormer mine_seg_sat/models/segformer.py unverified MIT (permissive) · 0872c74a03511111 · report
SegformerEncoder mine_seg_sat/models/segformer.py unverified MIT (permissive) · e440e08cb0e8c48a · report
sithu31296/semantic-segmentation community (archive-listed) 5 samples, 4 ran (0 instrument-failed) · 1 unverified
ConvModule in 8 papers semseg/models/segformer.py ran fingerprinted MIT (permissive) · 8e948e9bc7b5fc17 · report fixture: returned ConvModule · level recorded 2026-09-22
MLP in 375 papers semseg/models/segformer.py ran MIT (permissive) · 4e9019271c54ae21 · report fixture: returned MLP · level recorded 2026-09-22
SegFormer semseg/models/segformer.py ran MIT (permissive) · 21d890715e16872d · report fixture: returned SegFormer · level recorded 2026-09-22
SegFormerHead semseg/models/segformer.py ran MIT (permissive) · d8c46d51ed5de622 · report fixture: returned SegFormerHead · level recorded 2026-09-22
BaseModel in 92 papers semseg/models/segformer.py unverified MIT (permissive) · c1b48d941aafb38e · report
DavidLandup0/deepvision community (archive-listed) 7 samples, 4 ran (2 instrument-failed) · 3 unverified
SegFormerHead deepvision/models/segmentation/segformer/segformer.py ran · our draft was wrong Apache-2.0 (permissive) · 68f9e28fcc23bfc1 · report fixture: signature in_dims:list/1/int, num_classes:int/0/int, backend:str/0/str, embed_dim:int/0/int, name:str/0/str · returned __SegFormerHeadPT · level recorded 2026-09-22
__SegFormerHeadPT deepvision/models/segmentation/segformer/segformer.py ran · metamorphic tier: well formed Apache-2.0 (permissive) · df798977d45ef10d · report fixture: signature features:tuple/1/Tensor · returned Tensor · level recorded 2026-09-22
__SegFormerHeadTF deepvision/models/segmentation/segformer/segformer.py ran Apache-2.0 (permissive) · d1541cb3c3105894 · report fixture: signature features:list/1/EagerTensor · returned EagerTensor · level recorded 2026-09-24
parse_model_inputs deepvision/models/segmentation/segformer/segformer.py ran; the contract check could not run (RAISES) Apache-2.0 (permissive) · 49d6368d9a869361 · report fixture: signature backend:str/0/str, input_shape:tuple/1/int, input_tensor:NoneType/0/? · returned KerasTensor · level recorded 2026-09-22
SegFormerB0 deepvision/models/segmentation/segformer/segformer.py unverified Apache-2.0 (permissive) · d4b6a3c4e865a3cf · report
__SegFormerPT deepvision/models/segmentation/segformer/segformer.py unverified Apache-2.0 (permissive) · b83489fb3cb1ed11 · report
__SegFormerTF deepvision/models/segmentation/segformer/segformer.py unverified Apache-2.0 (permissive) · c42717f8aa0ef841 · report
JoegameZhou/Segmenter_workers community (archive-listed) 6 samples, 6 ran (2 instrument-failed) · 0 unverified
arg_to_varname in 10 papers src/configs/parser.py ran · our draft was wrong fingerprinted Apache-2.0 (permissive) here; the stored copy is pointer only · 177a4abf40724f6e · report fixture: signature st:str/0/str · returned str · level recorded 2026-09-25
argv_to_vars in 10 papers src/configs/parser.py ran Apache-2.0 (permissive) · 0355ab2cd3f47412 · report fixture: signature argv:list/1/str · returned list · level recorded 2026-09-24
assignment src/models/layers/weight_init.py ran Apache-2.0 (permissive) · 615dd6b5188585cd · report fixture: signature arr:ndarray/0/float, num:float/0/float · returned ndarray · level recorded 2026-09-24
calculate_gain in 14 papers src/models/layers/weight_init.py ran Apache-2.0 (permissive) · 882f145326210430 · report fixture: signature nonlinearity:str/0/str, param:NoneType/0/? · returned float · level recorded 2026-09-24
kaiming_uniform in 5 papers src/models/layers/weight_init.py ran Apache-2.0 (permissive) · 039303a649214f32 · report fixture: signature arr:ndarray/2/float, a:float/0/float, mode:str/0/str, nonlinearity:str/0/str · returned ndarray · level recorded 2026-09-24
trim_preceding_hyphens in 10 papers src/configs/parser.py ran · our draft was wrong fingerprinted Apache-2.0 (permissive) here; the stored copy is pointer only · a99f170ebd682af5 · report fixture: signature st:str/0/str · returned str · level recorded 2026-09-25
rulixiang/segformer-pytorch community 2 samples, 2 ran (2 instrument-failed) · 0 unverified
cal_eta dist_train.py ran · our draft was wrong no licence file found · pointer only · 68389231090ed0d4 · report fixture: signature time0:datetime/0/?, cur_iter:int/0/int, total_iter:int/0/int · returned tuple · level recorded, date not recorded
scores in 12 papers utils/eval_seg.py ran · our draft was wrong no licence file found · pointer only · 771525e476e6dcb2 · report fixture: signature label_trues:list/1/ndarray, label_preds:list/1/ndarray, num_classes:int/0/int · returned dict · level recorded, date not recorded
Repository not located 1 sample, 1 ran (0 instrument-failed) · 0 unverified
1 sample is 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.
cast_tuple in 58 papers repository not located ran · violated contract fingerprinted licence of this copy not recorded · 1f7c550e65142ef8 · report fixture: signature val:int/0/int, depth:int/0/int · returned tuple · level recorded 2026-09-22

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.

6 assemblies on this paper: 6 harvested, 0 generated, listed in 1 groups (siblings lifted in one batch form one group).

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

harvested Assembly for SegFormer (graph method: SegFormer). 12 bricks; individually on a fixture: 9 ran (of which 8 constructed an object rather than computing a result; 9 with no instrument failure: 0 honoured, 0 violated, 9 with no contract checked; 0 where Syntology's instrument failed) · 3 unverified. The assembly has not been run end-to-end.

one of 6 siblings

  1. SegFormer_B0 entry · level 0 · unverified · 41a9c09a3f658a00
  2. MixVisionTransformer model · level 0 · unverified · d47ffdc9c58366e1
  3. ResizeLayer helper · level 0 · unverified · 51b3f94442e3e993
  4. SegFormerHead model · level 1 · ran · bc90afe01ade811b
  5. Block helper · level 1 · ran · 0730c486d30ba0ad
  6. OverlapPatchEmbed helper · level 1 · ran · 31c432acc4aa0d25
  7. ConvModule helper · level 1 · ran · 985e0029300d8b09
  8. MLP helper · level 1 · ran · 7ff5661a4c26fcb7
  9. Attention helper · level 1 · ran · c26b3dcb83805d8b
  10. DropPath helper · level 1 · ran · 4b04a139b83d832e
  11. Mlp helper · level 1 · ran · 635ca5ddeb4973f7
  12. DWConv helper · level 1 · ran · 4817e9583fbf6b3c

harvested Assembly for SegFormer (graph method: SegFormer). 10 bricks; individually on a fixture: 6 ran (of which 6 constructed an object rather than computing a result; 6 with no instrument failure: 0 honoured, 0 violated, 6 with no contract checked; 0 where Syntology's instrument failed) · 4 unverified; every one of the 6 samples that ran constructed an object rather than computing a result. The assembly has not been run end-to-end.

one of 6 siblings

  1. SegFormer entry · level 0 · unverified · d2ba4808c201185e
  2. MixTransformer model · level 0 · unverified · e5084c52a530491b
  3. SegFormerHead model · level 1 · ran · 6caaf2219052f191
  4. Block model · level 0 · unverified · d3ae12f830ae7473
  5. OverlapPatchEmbed helper · level 1 · ran · df41558bed765430
  6. MLPHead model · level 1 · ran · 1cca7df806bbd21f
  7. DropPath helper · level 1 · ran · 8085cbc641b30993
  8. EfficientSelfAttention model · level 0 · unverified · 1bfaa1f3894b09c5
  9. TransformerMLP helper · level 1 · ran · 6afe61dba51eef77
  10. DepthwiseConv2D helper · level 1 · ran · ccadeb436e01c85c

harvested Assembly for SegFormer (graph method: SegFormer). 5 bricks; individually on a fixture: 4 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; 0 where Syntology's instrument failed) · 1 unverified; every one of the 4 samples that ran constructed an object rather than computing a result. The assembly has not been run end-to-end.

one of 6 siblings

  1. SegFormer entry · level 1 · ran · 21d890715e16872d
  2. BaseModel helper · level 0 · unverified · c1b48d941aafb38e
  3. SegFormerHead model · level 1 · ran · d8c46d51ed5de622
  4. ConvModule helper · level 1 · ran · 8e948e9bc7b5fc17
  5. MLP model · level 1 · ran · 4e9019271c54ae21

harvested Assembly for SegFormer (graph method: SegFormer). 8 bricks; individually on a fixture: 8 ran (of which 6 constructed an object rather than computing a result; 8 with no instrument failure: 0 honoured, 1 violated, 7 with no contract checked; 0 where Syntology's instrument failed) · 0 unverified. The assembly has not been run end-to-end.

one of 6 siblings

  1. Segformer entry · level 1 · ran · ce46b8418261110a
  2. MiT model · level 1 · ran · a3c01f13dc01b670
  3. cast_tuple helper · level 1 · ran · violated contract · 1f7c550e65142ef8
  4. EfficientSelfAttention model · level 1 · ran · 542396928de7e36a
  5. MixFeedForward model · level 1 · ran · ff4bb124dc115d63
  6. PreNorm model · level 1 · ran · 748000c608a20c77
  7. DsConv2d model · level 1 · ran · 10e1daaea8547fb9
  8. LayerNorm model · level 1 · ran · f919bd05baa8dbfe

harvested Assembly for SegFormer (graph method: SegFormer). 7 bricks; individually on a fixture: 4 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; 2 where Syntology's instrument failed) · 3 unverified. The assembly has not been run end-to-end.

one of 6 siblings

  1. SegFormerB0 entry · level 0 · unverified · d4b6a3c4e865a3cf
  2. SegFormerHead model · level 1 · ran · our draft was wrong · 68f9e28fcc23bfc1
  3. __SegFormerHeadPT helper · level 1 · ran · df798977d45ef10d
  4. __SegFormerHeadTF helper · level 1 · ran · d1541cb3c3105894
  5. __SegFormerPT helper · level 0 · unverified · b83489fb3cb1ed11
  6. __SegFormerTF helper · level 0 · unverified · c42717f8aa0ef841
  7. parse_model_inputs helper · level 1 · ran; the contract check could not run (RAISES) · 49d6368d9a869361

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

one of 6 siblings

  1. SegFormer entry · level 0 · unverified · 0872c74a03511111
  2. SegFormerDecoder model · level 1 · ran · c1cec60a6c890710
  3. SegformerEncoder model · level 0 · unverified · e440e08cb0e8c48a
  4. OverlapPatchEmbedding model · level 1 · ran · 79ded40dd7812caf
  5. TransformerBlock model · level 1 · ran · 89979e961601b054
  6. EfficientSelfAttention model · level 1 · ran · 5049266ef99b4d26
  7. MixedFeedForwardNetwork model · level 1 · ran · 46464f58f54aa5a0

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="2105.15203") returns these assemblies with each brick's role and wiring (how to connect).

Tasks

2D Semantic SegmentationC++ codeCrack SegmentationDecoderSemantic SegmentationThermal Image Segmentation

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
2D Semantic Segmentation WildScenes Segformer (MiT-B5) mIoU 40.83 #5 of 5 Archive leaderboard report
Crack Segmentation CrackVision12K SegFormer mIoU 0.57969 #4 of 4 Archive leaderboard report
Semantic Segmentation ADE20K SegFormer-B5 Params (M) 84.7 #92 of 235 Archive leaderboard report
Semantic Segmentation ADE20K SegFormer-B5 Validation mIoU 51.8 #92 of 235 Archive leaderboard report
Semantic Segmentation ADE20K SegFormer-B4 Params (M) 64.1 #101 of 235 Archive leaderboard report
Semantic Segmentation ADE20K SegFormer-B4 Validation mIoU 51.1 #101 of 235 Archive leaderboard report
Semantic Segmentation ADE20K SegFormer-B0 Params (M) 3.8 #224 of 235 Archive leaderboard report
Semantic Segmentation ADE20K SegFormer-B0 Validation mIoU 37.4 #224 of 235 Archive leaderboard report
Semantic Segmentation ADE20K val SegFormer-B5(MS, 87M #Params, ImageNet-1K pretrain) mIoU 51.8 #44 of 95 Archive leaderboard report
Semantic Segmentation COCO-Stuff full SegFormer-B5 (Single Scale) Mean IoU (class) 46.7 #1 of 1 Archive leaderboard report
Semantic Segmentation Cityscapes test SegFormer (MiT-B5, Mapillary) Mean IoU (class) 83.1% #19 of 105 Archive leaderboard report
Semantic Segmentation Cityscapes val SegFormer (MiT-B5, Mapillary) mIoU 84.0 #21 of 99 Archive leaderboard report
Semantic Segmentation Cityscapes val SegFormer-B0 Validation mIoU 76.2 #99 of 99 Archive leaderboard report
Semantic Segmentation DADA-seg SegFormer (MiT-B3) mIoU 27.0 #10 of 28 Archive leaderboard report
Semantic Segmentation DADA-seg SegFormer (MiT-B2) mIoU 21.2 #18 of 28 Archive leaderboard report
Semantic Segmentation DADA-seg SegFormer (MiT-B1) mIoU 16.6 #26 of 28 Archive leaderboard report
Semantic Segmentation DDD17 SegFormer-B2 mIoU 71.05 #5 of 9 Archive leaderboard report
Semantic Segmentation DELIVER SegFormer mIoU 57.20 #9 of 9 Archive leaderboard report
Semantic Segmentation DSEC SegFormer-B2 mIoU 71.99 #4 of 9 Archive leaderboard report
Semantic Segmentation DensePASS SegFormer (MiT-B2) mIoU 42.4% #12 of 36 Archive leaderboard report
Semantic Segmentation DensePASS SegFormer (MiT-B1) mIoU 38.5% #16 of 36 Archive leaderboard report
Semantic Segmentation EventScape SegFormer-B4 mIoU 59.86 #3 of 12 Archive leaderboard report
Semantic Segmentation EventScape SegFormer-B2 mIoU 58.69 #4 of 12 Archive leaderboard report
Semantic Segmentation Fine-Grained Grass Segmentation Dataset SegFormer mIoU 48.29 #4 of 10 Archive leaderboard report
Semantic Segmentation Potsdam SegFormer-B2 mIoU 84.65 #5 of 11 Archive leaderboard report
Semantic Segmentation Potsdam SegFormer-B1 mIoU 84.37 #6 of 11 Archive leaderboard report
Semantic Segmentation Potsdam SegFormer-B0 mIoU 83.67 #9 of 11 Archive leaderboard report
Semantic Segmentation SELMA SegFormer mIoU 77.2 #2 of 7 Archive leaderboard report
Semantic Segmentation SpectralWaste SegFormer (HYPER) mIoU 54.3 #3 of 8 Archive leaderboard report
Semantic Segmentation SpectralWaste SegFormer (HYPER3) mIoU 53.5 #4 of 8 Archive leaderboard report
Semantic Segmentation SpectralWaste SegFormer (RGB) mIoU 48.4 #7 of 8 Archive leaderboard report
Semantic Segmentation SynPASS SegFomrer mIoU 37.24% #3 of 6 Archive leaderboard report
Semantic Segmentation Synthetic Bathing Perception SegFormer mIoU 86.86 #5 of 5 Archive leaderboard report
Semantic Segmentation UPLight SegFormer-B2 (RGB) mIoU 89.60 #4 of 8 Archive leaderboard report
Semantic Segmentation US3D SegFormer-B2 mIoU 75.14 #6 of 11 Archive leaderboard report
Semantic Segmentation US3D SegFormer-B1 mIoU 74.19 #8 of 11 Archive leaderboard report
Semantic Segmentation US3D SegFormer-B0 mIoU 71.80 #11 of 11 Archive leaderboard report
Semantic Segmentation UrbanLF SegFormer mIoU (Real) 82.20 #8 of 14 Archive leaderboard report
Semantic Segmentation UrbanLF SegFormer mIoU (Syn) 78.53 #8 of 14 Archive leaderboard report
Semantic Segmentation Vaihingen SegFormer-B1 mIoU 76.92 #6 of 13 Archive leaderboard report
Semantic Segmentation Vaihingen SegFormer-B2 mIoU 76.69 #9 of 13 Archive leaderboard report
Semantic Segmentation Vaihingen SegFormer-B0 mIoU 75.57 #11 of 13 Archive leaderboard report
Semantic Segmentation ZJU-RGB-P SegFormer-B2 (RGB) mIoU 89.6 #9 of 13 Archive leaderboard report
Thermal Image Segmentation MFN Dataset SegFormer (B4) mIOU 54.8 #36 of 55 Archive leaderboard report
Thermal Image Segmentation MFN Dataset SegFormer (B2) mIOU 53.2 #38 of 55 Archive leaderboard report
Thermal Image Segmentation RGB-T-Glass-Segmentation SegFormer MAE 0.053 #11 of 22 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: Mix-FFN, SegFormer

Absolute Position EncodingsAdamAttentionBPEConvolutionDense ConnectionsDropoutLabel SmoothingLayer NormalizationLinear LayerMix-FFNMulti-Head AttentionPosition-Wise Feed-Forward LayerResidual ConnectionSegFormerSoftmaxTransformer

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