SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · aaai_20752 · AAAI · 2022

Provable Guarantees for Understanding Out-of-Distribution Detection

Yixuan Li, Peyman Morteza

Open in the Atlas

Code that ran

14 of 14 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; 1 honoured the contract we drafted, 0 violated it and 7 have no contract verdict; 6 failures of our instrument; 0 no recorded run), measured 2026-11-22; 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/aaai_20752.json · badge: /api/ran/aaai_20752.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
PeymanMorteza/GEM canonical (classifier label)
classifier basis: census
14 ran · 0 no recorded run
FunctionStatusWhere it lives
ODIN Ran
ran; our contract draft was wrong, not the code
PeymanMorteza/GEM/utils/score_calculation.py
code served (permissive licence) · get_code("17bbd961aaf86569")
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 ndarray, ndim 2, dtype family float, shape (2, 3) as a later run recorded it. The image it ran in was not recorded. The code mentions cuda, and the harness aliases cuda to the CPU for every sample; whether this run needed that is not recorded. If your copy returns the same type, 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
from torch.autograd import Variable
import torch.nn as nn
import numpy as np

torch.manual_seed(0)

# Build a minimal model that takes input and produces outputs
class SimpleModel(nn.Module):
    def __init__(self, input_dim, num_classes):
        super(SimpleModel, self).__init__()
        self.fc = nn.Linear(input_dim, num_classes)
    
    def forward(self, x):
        return self.fc(x)

# Create model
model = SimpleModel(input_dim=4, num_classes=3)

# Create input tensor with requires_grad=True for backward pass
batch_size = 2
input_dim = 4
num_classes = 3
inputs = Variable(torch.randn(batch_size, input_dim), requires_grad=True)

# Create outputs from model
outputs = model(inputs)

# Parameters for ODIN function
temper = 1.0
noiseMagnitude1 = 0.1

ARGS = {
    "inputs": inputs,
    "outputs": outputs,
    "model": model,
    "temper": temper,
    "noiseMagnitude1": noiseMagnitude1
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 2, dtype family float, shape (2, 3) — shape (2, 3) 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
calib_err Ran
ran on a synthesized input
PeymanMorteza/GEM/utils/calibration_tools.py
code served (permissive licence) · get_code("509aa629e46cf950")
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 float32, 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 numpy as np

# Create small test data for calibration error calculation
np.random.seed(0)

# Generate confidence scores (should be between 0 and 1)
confidence = np.array([0.9, 0.8, 0.7, 0.6, 0.5, 0.4, 0.3, 0.2, 0.1, 0.95], dtype=np.float32)

# Generate correctness labels (binary: 0 or 1)
correct = np.array([1, 1, 0, 1, 0, 0, 1, 0, 1, 1], dtype=np.float32)

# Define arguments
ARGS = {
    'confidence': confidence,
    'correct': correct,
    'p': '2',
    'beta': 3
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type float32, 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
fpr_and_fdr_at_recall Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
PeymanMorteza/GEM/utils/display_results.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from Wuyingwen/Pursuing-Feature-Separation-for-OOD-Detection, is pointer only (licence: NONE) · get_code("7242da839b9822be")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from Wuyingwen/Pursuing-Feature-Separation-for-OOD-Detection, and GitHub detected no licence in that repository when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

get_GEM_Mahalanobis_score Ran
ran on a synthesized input
PeymanMorteza/GEM/utils/score_calculation.py
code served (permissive licence) · get_code("2fd8a4b75750317c")
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 tuple. The image it ran in was not recorded. The code mentions cuda, and the harness aliases cuda to the CPU for every sample; whether this run needed that is not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn
import numpy as np
from torch.utils.data import DataLoader, TensorDataset

torch.manual_seed(0)

# Create a simple model with intermediate_forward method
class SimpleModel(nn.Module):
    def __init__(self):
        super(SimpleModel, self).__init__()
        self.conv1 = nn.Conv2d(3, 8, kernel_size=3, padding=1)
        self.conv2 = nn.Conv2d(8, 16, kernel_size=3, padding=1)
        self.fc = nn.Linear(16 * 4 * 4, 10)
        
    def intermediate_forward(self, x, layer_index):
        if layer_index == 0:
            x = self.conv1(x)
            return x
        elif layer_index == 1:
            x = self.conv1(x)
            x = self.conv2(x)
            return x
        else:
            x = self.conv1(x)
            x = self.conv2(x)
            x = torch.nn.functional.adaptive_avg_pool2d(x, (4, 4))
            x = x.view(x.size(0), -1)
            x = self.fc(x)
            return x
    
    def forward(self, x):
        return self.intermediate_forward(x, 2)

# Create model
model = SimpleModel()

# Create test data loader
batch_size = 2
num_classes = 3
num_samples = 4
feature_dim = 8
height, width = 4, 4

# Create dummy image data (3 channels, 4x4 images)
x_test = torch.randn(num_samples, 3, height, width)
y_test = torch.randint(0, num_classes, (num_samples,))
test_dataset = TensorDataset(x_test, y_test)
test_loader = DataLoader(test_dataset, batch_size=batch_size)

# Create sample_mean: list indexed by layer_index, each containing tensor of shape (num_classes, feature_dim)
layer_index = 0
sample_mean = [torch.randn(num_classes, feature_dim), torch.randn(num_classes, feature_dim)]

# Create precision matrices: list indexed by layer_index, each containing tensor of shape (feature_dim, feature_dim)
precision = [torch.eye(feature_dim), torch.eye(feature_dim)]

# Function parameters
ARGS = {
    'model': model,
    'test_loader': test_loader,
    'num_classes': num_classes,
    'sample_mean': sample_mean,
    'precision': precision,
    'layer_index': layer_index,
    'magnitude': 0.002,
    'num_batches': 2,
    'in_dist': False,
    'GEM': 0
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple — a tuple output has no array shape of its own; none is 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
get_measures Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
PeymanMorteza/GEM/utils/display_results.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from Wuyingwen/Pursuing-Feature-Separation-for-OOD-Detection, is pointer only (licence: NONE) · get_code("b9b6c36f7a3e7e67")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from Wuyingwen/Pursuing-Feature-Separation-for-OOD-Detection, and GitHub detected no licence in that repository when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

get_ood_scores_odin Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
PeymanMorteza/GEM/utils/score_calculation.py
code served (permissive licence) · get_code("3ad01de2f86dccba")
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 ndarray, ndim 1, dtype family float, shape (4,) as a later run recorded it. The image it ran in was not recorded. The code mentions cuda, and the harness aliases cuda to the CPU for every sample; whether this run needed that is not recorded. If your copy returns the same type, 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
import numpy as np

torch.manual_seed(0)

# Create a simple model
class SimpleNet(nn.Module):
    def __init__(self):
        super(SimpleNet, self).__init__()
        self.fc = nn.Linear(4, 3)
    
    def forward(self, x):
        return self.fc(x)

# Create model instance
net = SimpleNet()
net.eval()

# Create a simple data loader (list of batches)
# Each batch is (data, target)
data1 = torch.randn(2, 4)
target1 = torch.tensor([0, 1])
data2 = torch.randn(2, 4)
target2 = torch.tensor([2, 0])
loader = [(data1, target1), (data2, target2)]

# Define parameters
bs = 2  # batch size
ood_num_examples = 4
T = 1.0  # temperature
noise = 0.1  # noise magnitude
in_dist = False

ARGS = {
    'loader': loader,
    'net': net,
    'bs': bs,
    'ood_num_examples': ood_num_examples,
    'T': T,
    'noise': noise,
    'in_dist': in_dist
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 1, dtype family float, shape (4,) — shape (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
mahalanobis Ran
ran on a synthesized input
PeymanMorteza/GEM/simul/GEM.py
code served (permissive licence) · get_code("143ca50e74c47d4e")
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 float64, ndim 0, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim 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 numpy as np

np.random.seed(0)

x = np.array([1.0, 2.0, 3.0, 4.0])
mu = np.array([0.5, 1.5, 2.5, 3.5])
phi = 2.0

ARGS = {
    "x": x,
    "mu": mu,
    "phi": phi,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type float64, ndim 0, dtype family float — no output shape is stated: observe_signature.py recorded float ndim 0, which disagrees with the fixture run
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
mean_generator Ran
ran on a synthesized input
PeymanMorteza/GEM/simul/GEM.py
code served (permissive licence) · get_code("ce38aebfc39a1578")
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 list, ndim 2, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim 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 numpy as np

ARGS = {
    "d": 8,
    "k": 2,
    "beta": 0.5,
    "normalize": 0
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type list, ndim 2, dtype family float — a list output has no array shape of its own; none is stated
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
rescaled_GEM_score Ran
ran on a synthesized input
PeymanMorteza/GEM/simul/GEM.py
code served (permissive licence) · get_code("0b73df80b5382f05")
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 float64, ndim 0, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim 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 numpy as np

# Create small numpy arrays for testing
np.random.seed(0)

# x: a single data point (feature vector)
x = np.array([1.0, 2.0, 3.0, 4.0])

# mean: list of numpy arrays representing in-distribution means
# Create 3 different mean vectors, each of dimension 4
mean = [
    np.array([0.5, 1.5, 2.5, 3.5]),
    np.array([1.2, 2.2, 3.2, 4.2]),
    np.array([0.8, 1.8, 2.8, 3.8])
]

# phi: float magnitude of covariance
phi = 0.5

ARGS = {
    "x": x,
    "mean": mean,
    "phi": phi
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type float64, ndim 0, dtype family float — no output shape is stated: observe_signature.py recorded float ndim 0, which disagrees with the fixture run
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
soft_f1 Ran
ran on a synthesized input
PeymanMorteza/GEM/utils/calibration_tools.py
code served (permissive licence) · get_code("0a958e667c8c24c8")
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

torch.manual_seed(0)

# confidence: tensor of confidence scores (probabilities)
# correct: tensor of binary correctness labels (0 or 1)
# Both should be the same shape for element-wise operations

confidence = torch.tensor([0.9, 0.3, 0.7, 0.2], dtype=torch.float32)
correct = torch.tensor([1.0, 1.0, 0.0, 0.0], dtype=torch.float32)

ARGS = {
    "confidence": confidence,
    "correct": correct,
}
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
stable_cumsum Ran
ran, honoured the contract we drafted
PeymanMorteza/GEM/utils/display_results.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from fanghenshaometeor/ood-kernel-pca-ext, is pointer only (licence: NONE) · get_code("d4acb3120a027622")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from fanghenshaometeor/ood-kernel-pca-ext, and GitHub detected no licence in that repository when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

tune_temp Ran
ran on a synthesized input
PeymanMorteza/GEM/utils/calibration_tools.py
code served (permissive licence) · get_code("3c09c6a69241b742")
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 float, 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 numpy as np
import torch

torch.manual_seed(0)

# Create small logits and labels for testing
# logits: shape (2, 4) - 2 samples, 4 classes
logits = np.array([
    [2.0, 1.0, 0.5, 0.1],
    [0.1, 3.0, 1.5, 0.2]
], dtype=np.float32)

# labels: shape (2,) - class indices for each sample
labels = np.array([0, 1], dtype=np.int64)

ARGS = {
    "logits": logits,
    "labels": labels,
    "binary_search": True,
    "lower": 0.2,
    "upper": 5.0,
    "eps": 0.0001
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
cvxpy, numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type float, 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
validation_split Ran
ran; our contract draft was wrong, not the code
PeymanMorteza/GEM/utils/validation_dataset.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from tmlr-group/ATOL, is pointer only (licence: NONE) · get_code("cb20792068df7c3c")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from tmlr-group/ATOL, and GitHub detected no licence in that repository when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

validation_split_folder Ran
ran; our contract draft was wrong, not the code
PeymanMorteza/GEM/utils/validation_dataset.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from tmlr-group/ATOL, is pointer only (licence: NONE) · get_code("3ab1a759c0161341")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from tmlr-group/ATOL, and GitHub detected no licence in that repository when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

Also present in other papers' code

Of 14 samples lifted from this paper's linked repositories, 14 carry a function identity and 0 do not. Of the 14 distinct functions with an identity, 10 are also present in other papers whose code we hold (5 with at least one paper dated earlier, 0 the same month, 7 later) and 4 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
stable_cumsum 29 other papers
6 dated earlier, 20 later, 3 same year, order not determinable · 1 only as an assembly brick
1911.11132 (2019-11)
Scaling Out-of-Distribution Detection for Real-World Settings
2003.08440 (2020-03)
Synthesize then Compare: Detecting Failures and Anomalies for Semantic Segmentation
2104.08812 (2021-04)
Contrastive Out-of-Distribution Detection for Pretrained Transformers
and 26 more not named here
fpr_and_fdr_at_recall 8 other papers
3 dated earlier, 3 later, 2 same year, order not determinable
1911.11132 (2019-11)
Scaling Out-of-Distribution Detection for Real-World Settings
2003.08440 (2020-03)
Synthesize then Compare: Detecting Failures and Anomalies for Semantic Segmentation
2108.04562 (2021-08)
Deep Metric Learning for Open World Semantic Segmentation
and 5 more not named here
get_measures 8 other papers
3 dated earlier, 3 later, 2 same year, order not determinable
1911.11132 (2019-11)
Scaling Out-of-Distribution Detection for Real-World Settings
2003.08440 (2020-03)
Synthesize then Compare: Detecting Failures and Anomalies for Semantic Segmentation
2108.04562 (2021-08)
Deep Metric Learning for Open World Semantic Segmentation
and 5 more not named here
calib_err 4 other papers
3 dated earlier, 1 same year, order not determinable
1907.07174 (2019-07)
Natural Adversarial Examples
2105.07581 (2021-05)
Vision Transformers are Robust Learners
2109.05211 (2021-09)
RobustART: Benchmarking Robustness on Architecture Design and Training Techniques
and 1 more not named here
ODIN 3 other papers
1 later, 2 same year, order not determinable
2202.01197 (2022-02)
VOS: Learning What You Don't Know by Virtual Outlier Synthesis
2202.03299 (2022-02)
Training OOD Detectors in their Natural Habitats
2409.18205 (2024-09)
Bridging OOD Detection and Generalization: A Graph-Theoretic View
get_ood_scores_odin 3 other papers
1 later, 2 same year, order not determinable
2202.01197 (2022-02)
VOS: Learning What You Don't Know by Virtual Outlier Synthesis
2202.03299 (2022-02)
Training OOD Detectors in their Natural Habitats
2409.18205 (2024-09)
Bridging OOD Detection and Generalization: A Graph-Theoretic View
soft_f1 2 other papers
1 dated earlier, 1 same year, order not determinable
1907.07174 (2019-07)
Natural Adversarial Examples
2202.03299 (2022-02)
Training OOD Detectors in their Natural Habitats
validation_split 2 other papers
2 later
2311.03236 (2023-11)
Out-of-distribution Detection Learning with Unreliable Out-of-distribution Sources
2403.09053 (2024-03)
Towards a theory of model distillation
tune_temp 1 other paper
1 same year, order not determinable
2202.03299 (2022-02)
Training OOD Detectors in their Natural Habitats
validation_split_folder 1 other paper
1 later
2311.03236 (2023-11)
Out-of-distribution Detection Learning with Unreliable Out-of-distribution Sources

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

Out-of-distribution (OOD) detection is important for deploying machine learning models in the real world, where test data from shifted distributions can naturally arise. While a plethora of algorithmic approaches have recently emerged for OOD detection, a critical gap remains in theoretical understanding. In this work, we develop an analytical framework that characterizes and unifies the theoretical understanding for OOD detection. Our analytical framework motivates a novel OOD detection method for neural networks, GEM, which demonstrates both theoretical and empirical superiority. In particular, on CIFAR-100 as in-distribution data, our method outperforms a competitive baseline by 16.57% (FPR95). Lastly, we formally provide provable guarantees and comprehensive analysis of our method, underpinning how various properties of data distribution affect the performance of OOD detection.

For agents

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

get_harvested_code_for_paper("aaai_20752")
get_code_for_paper("aaai_20752")
have("aaai_20752")

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

curl https://syntology.ai/api/ran/aaai_20752.json

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

[![Syntology run record](https://syntology.ai/api/ran/aaai_20752.svg)](https://syntology.ai/paper/aaai_20752)

Connect an agent — have() is free.