SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2302.03015 · ICLR · 2023

Exploring and Exploiting Decision Boundary Dynamics for Adversarial Robustness

Micah Goldblum, Tom Goldstein, Furong Huang, Yanchao Sun, Yuancheng Xu

arXiv · PDF · Open in the Atlas

Code that ran

4 of 12 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (2 of them constructed an object rather than computing a result; 1 honoured the contract we drafted, 1 violated it and 2 have no contract verdict; 0 failures of our instrument; 8 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/2302.03015.json · badge: /api/ran/2302.03015.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
yuancheng-xu/dynamics-aware-robust-training alias (classifier label)
classifier basis: classifier_2025
4 ran · 8 no recorded run
FunctionStatusWhere it lives
Attack Ran
ran on a synthesized input · constructed an object; no result was computed
yuancheng-xu/dynamics-aware-robust-training/core/DyART/DyART_Linf.py
code served (permissive licence) · get_code("b5c1ada412a147f0")
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, Attack: 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)

# The Attack class __init__ only takes clip_min and clip_max
# These are bounds for clipping adversarial perturbations
# They should be scalars or array-like values representing min/max per dimension

clip_min = 0.0
clip_max = 1.0

ARGS = {
    'clip_min': clip_min,
    'clip_max': clip_max,
}
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 Attack — 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 Attack 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
FABAttack_scalar Ran
ran on a synthesized input · constructed an object; no result was computed
yuancheng-xu/dynamics-aware-robust-training/core/DyART/DyART_Linf.py
code served (permissive licence) · get_code("54c90b4b9c297c04")
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, FABAttack_scalar: the run constructed the object and did not call it, so no result was computed. 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

torch.manual_seed(0)

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)

model = SimpleModel()

ARGS = {
    'norm': 'Linf',
    'n_restarts': 1,
    'n_iter': 10,
    'eps': 0.3,
    'alpha_max': 0.1,
    'eta': 1.05,
    'beta': 0.9,
    'verbose': 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 FABAttack_scalar — 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 FABAttack_scalar 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
replicate_input Ran
ran, honoured the contract we drafted
yuancheng-xu/dynamics-aware-robust-training/core/DyART/DyART_Linf.py
code served (permissive licence) · get_code("4e78e41a8926dc30")
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 2, dtype family float, shape (2, 4) 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)

# Create a small tensor for the 'x' parameter
x = torch.randn(2, 4)

# Build the ARGS dictionary
ARGS = {
    'x': x
}
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 2, dtype family float, shape (2, 4) — shape (2, 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
smooth_max Ran
ran, violated the contract we drafted
yuancheng-xu/dynamics-aware-robust-training/core/DyART/DyART_Linf.py
code served (permissive licence) · get_code("50714c3786441e94")
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 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

torch.manual_seed(0)

# x: bs * (num_classes - 1), the logit values of classes that are not the true class
# bs = batch size = 2
# num_classes - 1 = 4
x = torch.randn(2, 4)

# beta: scaling factor for the smooth max operation
beta = 0.5

ARGS = {
    "x": x,
    "beta": beta,
}
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 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
DyART_loss_Linf Not yet run
unverified (harvested, no recorded run)
yuancheng-xu/dynamics-aware-robust-training/core/DyART/DyART_Linf.py
code served (permissive licence) · get_code("23fb5907f52402c5")
How it ran

no invocation recorded.

LabelMixin Not yet run
unverified (harvested, no recorded run)
yuancheng-xu/dynamics-aware-robust-training/core/DyART/DyART_Linf.py
code served (permissive licence) · get_code("8ed3a2aa6ae2a7ba")
How it ran

no invocation recorded.

find_bdr_FAB Not yet run
unverified (harvested, no recorded run)
yuancheng-xu/dynamics-aware-robust-training/core/DyART/DyART_Linf.py
code served (permissive licence) · get_code("1dbfdf8db17c1adf")
How it ran

no invocation recorded.

find_bdr_Linf Not yet run
unverified (harvested, no recorded run)
yuancheng-xu/dynamics-aware-robust-training/core/DyART/DyART_Linf.py
code served (permissive licence) · get_code("e406ff37047a4a9f")
How it ran

no invocation recorded.

func_soft_margin Not yet run
unverified (harvested, no recorded run)
yuancheng-xu/dynamics-aware-robust-training/core/DyART/DyART_Linf.py
code served (permissive licence) · get_code("15f8f0967b36aa06")
How it ran

no invocation recorded.

get_radius_weights Not yet run
unverified (harvested, no recorded run)
yuancheng-xu/dynamics-aware-robust-training/core/DyART/DyART_Linf.py
code served (permissive licence) · get_code("a1e687f30713373e")
How it ran

no invocation recorded.

h_prime Not yet run
unverified (harvested, no recorded run)
yuancheng-xu/dynamics-aware-robust-training/core/DyART/DyART_Linf.py
code served (permissive licence) · get_code("9f6fad065902cd84")
How it ran

no invocation recorded.

zero_gradients Not yet run
unverified (harvested, no recorded run)
yuancheng-xu/dynamics-aware-robust-training/core/DyART/DyART_Linf.py
code served (permissive licence) · get_code("9522fba05d4568aa")
How it ran

no invocation recorded.

Also present in other papers' code

Of 12 samples lifted from this paper's linked repositories, 12 carry a function identity and 0 do not. Of the 12 distinct functions with an identity, 1 are also present in other papers whose code we hold (1 with at least one paper dated earlier, 0 the same month, 1 later) and 11 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
replicate_input
brick of this paper's assembly: helper
4 other papers
1 dated earlier, 3 later
2110.09468 (2021-10)
Improving Robustness using Generated Data
2405.16260 (2024-05)
Enhancing Consistency-Based Image Generation via Adversarialy-Trained Classification and Energy-Based Discrimination
2406.13283 (2024-06)
Large-Scale Dataset Pruning in Adversarial Training through Data Importance Extrapolation
and 1 more not named here

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

The robustness of a deep classifier can be characterized by its margins: the decision boundary's distances to natural data points. However, it is unclear whether existing robust training methods effectively increase the margin for each vulnerable point during training. To understand this, we propose a continuous-time framework for quantifying the relative speed of the decision boundary with respect to each individual point. Through visualizing the moving speed of the decision boundary under Adversarial Training, one of the most effective robust training algorithms, a surprising moving-behavior is revealed: the decision boundary moves away from some vulnerable points but simultaneously moves closer to others, decreasing their margins. To alleviate these conflicting dynamics of the decision boundary, we propose Dynamics-Aware Robust Training (DyART), which encourages the decision boundary to engage in movement that prioritizes increasing smaller margins. In contrast to prior works, DyART directly operates on the margins rather than their indirect approximations, allowing for more targeted and effective robustness improvement. Experiments on the CIFAR-10 and Tiny-ImageNet datasets verify that DyART alleviates the conflicting dynamics of the decision boundary and obtains improved robustness under various perturbation sizes compared to the state-of-the-art defenses. Our code is available at https://github.com/ Yuancheng-Xu/Dynamics-Aware-Robust-Training.

For agents

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

get_harvested_code_for_paper("2302.03015")
get_code_for_paper("2302.03015")
have("2302.03015")

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

curl https://syntology.ai/api/ran/2302.03015.json

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

[![Syntology run record](https://syntology.ai/api/ran/2302.03015.svg)](https://syntology.ai/paper/2302.03015)

Connect an agent — have() is free.