SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2303.13846 · CVPR · 2023

Feature Separation and Recalibration for Adversarial Robustness

Jae Woo, Yoon, Yoonki Kim, Junsik Cho, Sung-Eui Jung

arXiv · PDF · Open in the Atlas

Code that ran

2 of 2 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 2 have no contract verdict; 0 failures of our instrument; 0 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/2303.13846.json · badge: /api/ran/2303.13846.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
wkim97/FSR community
Papers with Code archive
2 ran · 0 no recorded run
FunctionStatusWhere it lives
GumbelSigmoid Ran
ran on a synthesized input
wkim97/FSR/models/resnet_fsr.py
code served (permissive licence) · get_code("fd4971bc244014ab")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, class construct-then-call prompt, model not recorded) and returned Tensor, 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 torch
import torch.nn as nn

class GumbelSigmoid(nn.Module):

    def __init__(self, tau=1.0):
        super(GumbelSigmoid, self).__init__()
        self.tau = tau
        self.softmax = nn.Softmax(dim=1)
        self.p_value = 1e-08

    def forward(self, x, is_eval=False):
        r = 1 - x
        x = (x + self.p_value).log()
        r = (r + self.p_value).log()
        if not is_eval:
            x_N = torch.rand_like(x)
            r_N = torch.rand_like(r)
        else:
            x_N = 0.5 * torch.ones_like(x)
            r_N = 0.5 * torch.ones_like(r)
        x_N = -1 * (x_N + self.p_value).log()
        r_N = -1 * (r_N + self.p_value).log()
        x_N = -1 * (x_N + self.p_value).log()
        r_N = -1 * (r_N + self.p_value).log()
        x = x + x_N
        x = x / (self.tau + self.p_value)
        r = r + r_N
        r = r / (self.tau + self.p_value)
        x = torch.cat((x, r), dim=1)
        x = self.softmax(x)
        return x


INIT_ARGS = {
    'tau': 1.0
}

ARGS = {
    'x': torch.rand(2, 4),
    'is_eval': False
}

if __name__ == '__main__':
    model = GumbelSigmoid(**INIT_ARGS)
    output = model(**ARGS)
    print("Output shape:", output.shape)
    print("Output:", output)
Setup block drafted by
a language model, class construct-then-call prompt, model not recorded
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 — no output shape is stated: observe_signature.py observed this class with its constructor sizes canonicalised where it could, which the node does not record, so its shape is not stated as this setup block's
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: torch and numpy seeded to 0 by the harness immediately before the class was constructed from INIT_ARGS (fixture_check.py)
Separation Ran
ran on a synthesized input
wkim97/FSR/models/resnet_fsr.py
code served (permissive licence) · get_code("efd9fb7d329802da")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, class construct-then-call prompt, model not recorded) and returned tuple. 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

INIT_ARGS = {
    'size': (4, 8, 8),
    'num_channel': 16,
    'tau': 0.1
}

ARGS = {
    'feat': torch.randn(2, 4, 8, 8),
    'is_eval': False
}
Setup block drafted by
a language model, class construct-then-call prompt, model not recorded
Imports
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: torch and numpy seeded to 0 by the harness immediately before the class was constructed from INIT_ARGS (fixture_check.py)

Also present in other papers' code

Of 2 samples lifted from this paper's linked repositories, 2 carry a function identity and 0 do not. Of the 2 distinct functions with an identity, 0 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 0 later) and 2 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.

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

Deep neural networks are susceptible to adversarial attacks due to the accumulation of perturbations in the feature level, and numerous works have boosted model robustness by deactivating the non-robust feature activations that cause model mispredictions. However, we claim that these malicious activations still contain discriminative cues and that with recalibration, they can capture additional useful information for correct model predictions. To this end, we propose a novel, easy-to-plugin approach named Feature Separation and Recalibration (FSR) that recalibrates the malicious, non-robust activations for more robust feature maps through Separation and Recalibration. The Separation part disentangles the input feature map into the robust feature with activations that help the model make correct predictions and the non-robust feature with activations that are responsible for model mispredictions upon adversarial attack. The Recalibration part then adjusts the non-robust activations to restore the potentially useful cues for model predictions. Extensive experiments verify the superiority of FSR compared to traditional deactivation techniques and demonstrate that it improves the robustness of existing adversarial training methods by up to 8.57% with small computational overhead. Codes are available at https://github.com/wkim97/FSR.

For agents

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

get_harvested_code_for_paper("2303.13846")
get_code_for_paper("2303.13846")
have("2303.13846")

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

curl https://syntology.ai/api/ran/2303.13846.json

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

[![Syntology run record](https://syntology.ai/api/ran/2303.13846.svg)](https://syntology.ai/paper/2303.13846)

Connect an agent — have() is free.