SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2101.10979 · CVPR · 2021

Prototypical Pseudo Label Denoising and Target Structure Learning for Domain Adaptive Semantic Segmentation

Dong Chen, Bo Zhang, Fang Wen, Pan Zhang, Ting Zhang, Yong Wang

arXiv · PDF · Open in the Atlas

Code that ran

8 of 17 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (6 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 6 have no contract verdict; 2 failures of our instrument; 9 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/2101.10979.json · badge: /api/ran/2101.10979.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
microsoft/ProDA marked official in Papers with Code
Papers with Code archive
8 ran · 9 no recorded run
FunctionStatusWhere it lives
Bottleneck Ran
ran on a synthesized input · constructed an object; no result was computed
microsoft/ProDA/models/adaptation_modelv2.py
code served (permissive licence) · get_code("6020d23a07c55f29")
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, Bottleneck: 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)

# Define BatchNorm as a parameter
BatchNorm = nn.BatchNorm2d

# Define downsample as a sequential module that matches the expected output channels
# The Bottleneck with expansion=4 will output planes*4 channels
# If stride > 1, we need to downsample the input to match
downsample = nn.Sequential(
    nn.Conv2d(64, 256, kernel_size=1, stride=1, bias=False),
    nn.BatchNorm2d(256, affine=True)
)

ARGS = {
    'inplanes': 64,
    'planes': 64,
    'stride': 1,
    'dilation': 1,
    'downsample': downsample,
    'BatchNorm': BatchNorm
}
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 Bottleneck — 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 Bottleneck 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
CallbackContext Ran
ran on a synthesized input · constructed an object; no result was computed
microsoft/ProDA/models/adaptation_modelv2.py
code served (permissive licence) · get_code("0e7e99a49473a940")
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, CallbackContext: 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)

class CallbackContext(object):
    pass

ARGS = {}
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 CallbackContext — 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 CallbackContext 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
Classifier_Module2 Ran
ran on a synthesized input · constructed an object; no result was computed
microsoft/ProDA/models/adaptation_modelv2.py
code served (permissive licence) · get_code("7cfa9b72e3a4749d")
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, Classifier_Module2: 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 = {
    'inplanes': 64,
    'dilation_series': [6, 12, 18],
    'padding_series': [6, 12, 18],
    'num_classes': 19,
    'droprate': 0.1,
    'use_se': True
}
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 Classifier_Module2 — 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 Classifier_Module2 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
FCDiscriminator Ran
ran on a synthesized input · constructed an object; no result was computed
microsoft/ProDA/models/adaptation_modelv2.py
code served (permissive licence) · get_code("23029f57b971a963")
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, FCDiscriminator: 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)

# FCDiscriminator.__init__ takes two parameters:
# - inplanes: number of input channels
# - planes: base number of filters (default=64)

ARGS = {
    'inplanes': 3,
    'planes': 64,
}
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 FCDiscriminator — 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 FCDiscriminator 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
PolynomialLR Ran
ran on a synthesized input · constructed an object; no result was computed
microsoft/ProDA/models/adaptation_modelv2.py
code served (permissive licence) · get_code("8a42c7371906d184")
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, PolynomialLR: 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
from torch.optim import SGD

torch.manual_seed(0)

# Create a minimal model
class SimpleModel(nn.Module):
    def __init__(self):
        super(SimpleModel, self).__init__()
        self.linear = nn.Linear(4, 2)
    
    def forward(self, x):
        return self.linear(x)

model = SimpleModel()

# Create an optimizer
optimizer = SGD(model.parameters(), lr=0.1)

# Define ARGS dict with all parameters
ARGS = {
    'optimizer': optimizer,
    'max_iter': 100,
    'decay_iter': 1,
    'gamma': 0.9,
    'last_epoch': -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 PolynomialLR — 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 PolynomialLR 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
SEBlock Ran
ran on a synthesized input · constructed an object; no result was computed
microsoft/ProDA/models/adaptation_modelv2.py
code served (permissive licence) · get_code("332f8cc5d0d081e6")
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, SEBlock: 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)

class SEBlock(nn.Module):

    def __init__(self, inplanes, r=16):
        super(SEBlock, self).__init__()
        self.global_pool = nn.AdaptiveAvgPool2d((1, 1))
        self.se = nn.Sequential(nn.Linear(inplanes, inplanes // r), nn.ReLU(inplace=True), nn.Linear(inplanes // r, inplanes), nn.Sigmoid())

    def forward(self, x):
        xx = self.global_pool(x)
        xx = xx.view(xx.size(0), xx.size(1))
        se_weight = self.se(xx).unsqueeze(-1).unsqueeze(-1)
        return x.mul(se_weight)

# Create a test input tensor with shape (batch=2, channels=32, height=4, width=3)
test_input = torch.randn(2, 32, 4, 3)

ARGS = {
    'inplanes': 32,
    'r': 16
}
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 SEBlock — 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 SEBlock 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
cross_entropy2d Ran
ran; our contract draft was wrong, not the code
microsoft/ProDA/models/adaptation_modelv2.py
code served (permissive licence) · get_code("e48b895c94b5aa40")
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 0, dtype family float, shape () 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 as nn

torch.manual_seed(0)

# Create input tensor: (batch_size=2, num_classes=4, height=3, width=5)
input_tensor = torch.randn(2, 4, 3, 5)

# Create target tensor: (batch_size=2, height=3, width=5) with class indices
target_tensor = torch.randint(0, 4, (2, 3, 5))

# Create a weight tensor for classes (one weight per class)
weight_tensor = torch.tensor([1.0, 1.0, 1.0, 1.0])

ARGS = {
    'input': input_tensor,
    'target': target_tensor,
    'weight': weight_tensor,
    'size_average': True,
    'softmax_used': False,
    'reduction': 'mean',
    'cls_num_list': None
}
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 0, dtype family float, shape () — shape () 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
get_scheduler Ran
ran; our contract draft was wrong, not the code
microsoft/ProDA/models/adaptation_modelv2.py
code served (permissive licence) · get_code("7101765897c8d17b")
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 PolynomialLR. 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 torch.optim.lr_scheduler import _LRScheduler

torch.manual_seed(0)

class PolynomialLR(_LRScheduler):
    def __init__(self, optimizer, max_iter, decay_iter=1, gamma=0.9, last_epoch=-1):
        self.decay_iter = decay_iter
        self.max_iter = max_iter
        self.gamma = gamma
        super(PolynomialLR, self).__init__(optimizer, last_epoch)

    def get_lr(self):
        factor = (1 - self.last_epoch / float(self.max_iter)) ** self.gamma
        factor = max(factor, 0)
        return [base_lr * factor for base_lr in self.base_lrs]

def get_scheduler(optimizer, opt):
    scheduler = PolynomialLR(optimizer, max_iter=opt.train_iters, gamma=0.9)
    return scheduler

class SimpleModel(nn.Module):
    def __init__(self):
        super(SimpleModel, self).__init__()
        self.linear = nn.Linear(4, 3)
    
    def forward(self, x):
        return self.linear(x)

class OptConfig:
    def __init__(self, train_iters):
        self.train_iters = train_iters

model = SimpleModel()
optimizer = torch.optim.SGD(model.parameters(), lr=0.1)
opt_config = OptConfig(train_iters=100)

ARGS = {
    "optimizer": optimizer,
    "opt": opt_config
}
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 PolynomialLR — a PolynomialLR output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a PolynomialLR 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
CustomModel Not yet run
unverified (harvested, no recorded run)
microsoft/ProDA/models/adaptation_modelv2.py
code served (permissive licence) · get_code("097fa03acb194ec7")
How it ran

no invocation recorded.

DataParallelWithCallback Not yet run
unverified (harvested, no recorded run)
microsoft/ProDA/models/adaptation_modelv2.py
code served (permissive licence) · get_code("621eff113a744b84")
How it ran

no invocation recorded.

Deeplab Not yet run
unverified (harvested, no recorded run)
microsoft/ProDA/models/adaptation_modelv2.py
code served (permissive licence) · get_code("5b7b14dc1a126e30")
How it ran

no invocation recorded.

ResNet101 Not yet run
unverified (harvested, no recorded run)
microsoft/ProDA/models/adaptation_modelv2.py
code served (permissive licence) · get_code("724f27e7f6fe421c")
How it ran

no invocation recorded.

affine_sample Not yet run
unverified (harvested, no recorded run)
microsoft/ProDA/models/adaptation_modelv2.py
code served (permissive licence) · get_code("8abaace0beabbac9")
How it ran

no invocation recorded.

execute_replication_callbacks Not yet run
unverified (harvested, no recorded run)
microsoft/ProDA/models/adaptation_modelv2.py
code served (permissive licence) · get_code("7308ae70af0cb672")
How it ran

no invocation recorded.

feat_prototype_distance_module Not yet run
unverified (harvested, no recorded run)
microsoft/ProDA/models/adaptation_modelv2.py
code served (permissive licence) · get_code("c0389bd83130fd48")
How it ran

no invocation recorded.

freeze_bn Not yet run
unverified (harvested, no recorded run)
microsoft/ProDA/models/adaptation_modelv2.py
code served (permissive licence) · get_code("fa564b743e0f7020")
How it ran

no invocation recorded.

freeze_bn_func Not yet run
unverified (harvested, no recorded run)
microsoft/ProDA/models/adaptation_modelv2.py
code served (permissive licence) · get_code("692f2a64673cce1b")
How it ran

no invocation recorded.

Also present in other papers' code

Of 17 samples lifted from this paper's linked repositories, 17 carry a function identity and 0 do not. Of the 17 distinct functions with an identity, 16 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 16 later) and 1 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
Bottleneck
brick of this paper's assembly: helper
2 other papers
2 later · 2 only as an assembly brick
2203.09744 (2022-03)
Class-Balanced Pixel-Level Self-Labeling for Domain Adaptive Semantic Segmentation
2308.13236 (2023-08)
Black-box Unsupervised Domain Adaptation with Bi-directional Atkinson-Shiffrin Memory
CallbackContext
brick of this paper's assembly: helper
2 other papers
2 later · 2 only as an assembly brick
2203.09744 (2022-03)
Class-Balanced Pixel-Level Self-Labeling for Domain Adaptive Semantic Segmentation
2308.13236 (2023-08)
Black-box Unsupervised Domain Adaptation with Bi-directional Atkinson-Shiffrin Memory
Classifier_Module2
brick of this paper's assembly: helper
2 other papers
2 later · 2 only as an assembly brick
2203.09744 (2022-03)
Class-Balanced Pixel-Level Self-Labeling for Domain Adaptive Semantic Segmentation
2308.13236 (2023-08)
Black-box Unsupervised Domain Adaptation with Bi-directional Atkinson-Shiffrin Memory
DataParallelWithCallback
brick of this paper's assembly: helper
2 other papers
2 later · 2 only as an assembly brick
2203.09744 (2022-03)
Class-Balanced Pixel-Level Self-Labeling for Domain Adaptive Semantic Segmentation
2308.13236 (2023-08)
Black-box Unsupervised Domain Adaptation with Bi-directional Atkinson-Shiffrin Memory
Deeplab
brick of this paper's assembly: helper
2 other papers
2 later · 1 only as an assembly brick
2203.09744 (2022-03)
Class-Balanced Pixel-Level Self-Labeling for Domain Adaptive Semantic Segmentation
2308.13236 (2023-08)
Black-box Unsupervised Domain Adaptation with Bi-directional Atkinson-Shiffrin Memory
FCDiscriminator
brick of this paper's assembly: helper
2 other papers
2 later · 2 only as an assembly brick
2203.09744 (2022-03)
Class-Balanced Pixel-Level Self-Labeling for Domain Adaptive Semantic Segmentation
2308.13236 (2023-08)
Black-box Unsupervised Domain Adaptation with Bi-directional Atkinson-Shiffrin Memory
PolynomialLR
brick of this paper's assembly: helper
2 other papers
2 later · 2 only as an assembly brick
2203.09744 (2022-03)
Class-Balanced Pixel-Level Self-Labeling for Domain Adaptive Semantic Segmentation
2308.13236 (2023-08)
Black-box Unsupervised Domain Adaptation with Bi-directional Atkinson-Shiffrin Memory
ResNet101
brick of this paper's assembly: helper
2 other papers
2 later · 1 only as an assembly brick
2203.09744 (2022-03)
Class-Balanced Pixel-Level Self-Labeling for Domain Adaptive Semantic Segmentation
2308.13236 (2023-08)
Black-box Unsupervised Domain Adaptation with Bi-directional Atkinson-Shiffrin Memory
SEBlock
brick of this paper's assembly: helper
2 other papers
2 later · 2 only as an assembly brick
2203.09744 (2022-03)
Class-Balanced Pixel-Level Self-Labeling for Domain Adaptive Semantic Segmentation
2308.13236 (2023-08)
Black-box Unsupervised Domain Adaptation with Bi-directional Atkinson-Shiffrin Memory
affine_sample
brick of this paper's assembly: helper
2 other papers
2 later · 2 only as an assembly brick
2203.09744 (2022-03)
Class-Balanced Pixel-Level Self-Labeling for Domain Adaptive Semantic Segmentation
2308.13236 (2023-08)
Black-box Unsupervised Domain Adaptation with Bi-directional Atkinson-Shiffrin Memory
cross_entropy2d
brick of this paper's assembly: helper
2 other papers
2 later · 2 only as an assembly brick
2203.09744 (2022-03)
Class-Balanced Pixel-Level Self-Labeling for Domain Adaptive Semantic Segmentation
2308.13236 (2023-08)
Black-box Unsupervised Domain Adaptation with Bi-directional Atkinson-Shiffrin Memory
execute_replication_callbacks
brick of this paper's assembly: helper
2 other papers
2 later · 2 only as an assembly brick
2203.09744 (2022-03)
Class-Balanced Pixel-Level Self-Labeling for Domain Adaptive Semantic Segmentation
2308.13236 (2023-08)
Black-box Unsupervised Domain Adaptation with Bi-directional Atkinson-Shiffrin Memory
feat_prototype_distance_module
brick of this paper's assembly: model
2 other papers
2 later · 2 only as an assembly brick
2203.09744 (2022-03)
Class-Balanced Pixel-Level Self-Labeling for Domain Adaptive Semantic Segmentation
2308.13236 (2023-08)
Black-box Unsupervised Domain Adaptation with Bi-directional Atkinson-Shiffrin Memory
freeze_bn
brick of this paper's assembly: helper
2 other papers
2 later · 2 only as an assembly brick
2203.09744 (2022-03)
Class-Balanced Pixel-Level Self-Labeling for Domain Adaptive Semantic Segmentation
2308.13236 (2023-08)
Black-box Unsupervised Domain Adaptation with Bi-directional Atkinson-Shiffrin Memory
freeze_bn_func
brick of this paper's assembly: helper
2 other papers
2 later · 2 only as an assembly brick
2203.09744 (2022-03)
Class-Balanced Pixel-Level Self-Labeling for Domain Adaptive Semantic Segmentation
2308.13236 (2023-08)
Black-box Unsupervised Domain Adaptation with Bi-directional Atkinson-Shiffrin Memory
get_scheduler
brick of this paper's assembly: helper
2 other papers
2 later · 2 only as an assembly brick
2203.09744 (2022-03)
Class-Balanced Pixel-Level Self-Labeling for Domain Adaptive Semantic Segmentation
2308.13236 (2023-08)
Black-box Unsupervised Domain Adaptation with Bi-directional Atkinson-Shiffrin Memory

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

Self-training is a competitive approach in domain adaptive segmentation, which trains the network with the pseudo labels on the target domain. However inevitably, the pseudo labels are noisy and the target features are dispersed due to the discrepancy between source and target domains. In this paper, we rely on representative prototypes, the feature centroids of classes, to address the two issues for unsupervised domain adaptation. In particular, we take one step further and exploit the feature distances from prototypes that provide richer information than mere prototypes. Specifically, we use it to estimate the likelihood of pseudo labels to facilitate online correction in the course of training. Meanwhile, we align the prototypical assignments based on relative feature distances for two different views of the same target, producing a more compact target feature space. Moreover, we find that distilling the already learned knowledge to a self-supervised pretrained model further boosts the performance. Our method shows tremendous performance advantage over state-of-the-art methods. The code is available at https://github.com/microsoft/ProDA.

For agents

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

get_harvested_code_for_paper("2101.10979")
get_code_for_paper("2101.10979")
have("2101.10979")

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

curl https://syntology.ai/api/ran/2101.10979.json

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

[![Syntology run record](https://syntology.ai/api/ran/2101.10979.svg)](https://syntology.ai/paper/2101.10979)

Connect an agent — have() is free.