SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2006.15207 · 2020

ATOM: Robustifying Out-of-distribution Detection Using Outlier Mining

arXiv · PDF · Open in the Atlas

Code that ran

3 of 5 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (0 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 0 have no contract verdict; 3 failures of our instrument; 2 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/2006.15207.json · badge: /api/ran/2006.15207.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
jfc43/informative-outlier-mining canonical (classifier label)
classifier basis: classifier_2025
3 ran · 1 no recorded run
copy not recorded community
not in the paper's declared repositories
0 ran · 1 no recorded run
FunctionStatusWhere it lives
get_msp_score Ran
ran; our contract draft was wrong, not the code
jfc43/informative-outlier-mining/eval_ood_detection.py
code served (permissive licence) · get_code("8511dff685e73d9d")
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 (2,) 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
import numpy as np

torch.manual_seed(0)

# Create a simple model that takes inputs and outputs logits
class SimpleClassifier(nn.Module):
    def __init__(self, input_dim, num_classes):
        super().__init__()
        self.linear = nn.Linear(input_dim, num_classes)
    
    def forward(self, x):
        return self.linear(x)

# Create model instance
model = SimpleClassifier(input_dim=4, num_classes=3)
model.eval()

# Create sample inputs: batch size 2, feature dimension 4
inputs = torch.randn(2, 4)

# Create method_args dict (empty dict is valid since function doesn't use it)
method_args = {}

ARGS = {
    "inputs": inputs,
    "model": model,
    "method_args": method_args
}
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 (2,) — shape (2,) 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_rowl_score Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
jfc43/informative-outlier-mining/eval_ood_detection.py
code served (permissive licence) · get_code("3574d883f0468bc9")
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 (2,) 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
import torch.nn.functional as F

torch.manual_seed(0)

# Define a simple model that takes inputs and produces outputs
class SimpleModel(nn.Module):
    def __init__(self, input_dim, output_dim):
        super().__init__()
        self.linear = nn.Linear(input_dim, output_dim)
    
    def forward(self, x):
        return self.linear(x)

# Create model with input_dim=4, output_dim=5 (5 classes: 0-4, where class 4 is num_classes)
model = SimpleModel(input_dim=4, output_dim=5)
model.eval()

# Create input tensor: batch_size=2, feature_dim=4
inputs = torch.randn(2, 4)

# Create method_args dict with num_classes
method_args = {
    'num_classes': 4
}

# Define ARGS mapping all parameters
ARGS = {
    'inputs': inputs,
    'model': model,
    'method_args': method_args,
    'raw_score': False
}
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 ndarray, ndim 1, dtype family float, shape (2,) — shape (2,) 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_sofl_score Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
jfc43/informative-outlier-mining/eval_ood_detection.py
code served (permissive licence) · get_code("f3bc884f0d0d3994")
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 (2,) 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
import torch.nn.functional as F
import numpy as np

torch.manual_seed(0)

# Build a minimal model that takes inputs and outputs logits
class SimpleModel(nn.Module):
    def __init__(self, input_dim, output_dim):
        super().__init__()
        self.linear = nn.Linear(input_dim, output_dim)
    
    def forward(self, x):
        return self.linear(x)

# Create inputs: batch size 2, input dimension 4
inputs = torch.randn(2, 4)

# Create model: input_dim=4, output_dim=6 (to have classes beyond num_classes)
model = SimpleModel(input_dim=4, output_dim=6)

# Create method_args with num_classes=3 (so indices 3:6 will be summed)
method_args = {
    'num_classes': 3
}

ARGS = {
    'inputs': inputs,
    'model': model,
    'method_args': method_args
}
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 (2,) — shape (2,) 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
accuracy Not yet run
unverified (harvested, no recorded run)
this paper's copy was not recorded; identical code first harvested from snu-mllab/PuzzleMix
pointer only · get_code("f0c9a29156911331")
How it ran

The body of this sample is withheld: this paper's copy of the code was not recorded (its link names no repository or source path), so no licence was read for it. 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. No invocation recorded.

select_ood Not yet run
unverified (harvested, no recorded run)
jfc43/informative-outlier-mining/train_atom.py
code served (permissive licence) · get_code("1b6c040456d8424b")
How it ran

no invocation recorded.

Also present in other papers' code

Of 5 samples lifted from this paper's linked repositories, 5 carry a function identity and 0 do not. Of the 5 distinct functions with an identity, 4 are also present in other papers whose code we hold (1 with at least one paper dated earlier, 1 the same month, 4 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
accuracy 71 other papers
29 dated earlier, 5 the same month, 37 later · 2 only as an assembly brick
1505.05424 (2015-05)
Weight Uncertainty in Neural Networks
1705.07204 (2017-05)
Ensemble Adversarial Training: Attacks and Defenses
1802.10399 (2018-02)
Compressing Neural Networks using the Variational Information Bottleneck
and 68 more not named here
get_msp_score 3 other papers
3 later
2111.09805 (2021-11)
DICE: Leveraging Sparsification for Out-of-Distribution Detection
2303.13995 (2023-03)
LINe: Out-of-Distribution Detection by Leveraging Important Neurons
2411.14049 (2024-11)
Out-Of-Distribution Detection with Diversification (Provably)
get_rowl_score 2 other papers
2 later
2111.09805 (2021-11)
DICE: Leveraging Sparsification for Out-of-Distribution Detection
2303.13995 (2023-03)
LINe: Out-of-Distribution Detection by Leveraging Important Neurons
get_sofl_score 2 other papers
2 later
2111.09805 (2021-11)
DICE: Leveraging Sparsification for Out-of-Distribution Detection
2303.13995 (2023-03)
LINe: Out-of-Distribution Detection by Leveraging Important Neurons

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

Detecting out-of-distribution (OOD) inputs is critical for safely deploying deep learning models in an open-world setting. However, existing OOD detection solutions can be brittle in the open world, facing various types of adversarial OOD inputs. While methods leveraging auxiliary OOD data have emerged, our analysis on illuminative examples reveals a key insight that the majority of auxiliary OOD examples may not meaningfully improve or even hurt the decision boundary of the OOD detector, which is also observed in empirical results on real data. In this paper, we provide a theoretically motivated method, Adversarial Training with informative Outlier Mining (ATOM), which improves the robustness of OOD detection. We show that, by mining informative auxiliary OOD data, one can significantly improve OOD detection performance, and somewhat surprisingly, generalize to unseen adversarial attacks. ATOM achieves state-of-the-art performance under a broad family of classic and adversarial OOD evaluation tasks. For example, on the CIFAR-10 in-distribution dataset, ATOM reduces the FPR (at TPR 95%) by up to 57.99% under adversarial OOD inputs, surpassing the previous best baseline by a large margin.

For agents

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

get_harvested_code_for_paper("2006.15207")
get_code_for_paper("2006.15207")
have("2006.15207")

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

curl https://syntology.ai/api/ran/2006.15207.json

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

[![Syntology run record](https://syntology.ai/api/ran/2006.15207.svg)](https://syntology.ai/paper/2006.15207)

Connect an agent — have() is free.