SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2104.12609 · ICLR · 2021

Published at ICLR 2021 Workshop on Security and Safety in Machine Learning Systems PATCHGUARD++: EFFICIENT PROVABLE ATTACK DE-TECTION AGAINST ADVERSARIAL PATCHES

Prateek Mittal, Chong Xiang

arXiv · PDF · Open in the Atlas

Code that ran

1 of 1 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; 1 failure 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/2104.12609.json · badge: /api/ran/2104.12609.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
inspire-group/PatchGuard marked official in Papers with Code
Papers with Code archive
1 ran · 0 no recorded run
FunctionStatusWhere it lives
pg2_detection_provable Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
inspire-group/PatchGuard/utils/defense_utils.py
code served (permissive licence) · get_code("8547ff30cf40c4a7")
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) and returned int, ndim 0, dtype family int, 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

# Set random seed for reproducibility
np.random.seed(0)

# Create a small local_feature tensor
# Shape: [feature_size_x, feature_size_y, num_cls]
# Using distinct dimensions: 8x7x5
feature_size_x = 8
feature_size_y = 7
num_cls = 5

local_feature = np.random.randn(feature_size_x, feature_size_y, num_cls)

# Ensure the global prediction matches the label
# by making the label class have the highest mean value
label = 2
global_feature = np.mean(local_feature, axis=(0, 1))
global_feature[label] = np.max(global_feature) + 1.0
local_feature[:, :, label] += 2.0

# Detection threshold tau in [0, 1]
tau = 0.5

# Window shape
window_shape = [6, 6]

ARGS = {
    "local_feature": local_feature,
    "label": label,
    "tau": tau,
    "window_shape": window_shape
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
numpy, scipy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type int, ndim 0, dtype family int, 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

Also present in other papers' code

Of 1 samples lifted from this paper's linked repositories, 1 carry a function identity and 0 do not. Of the 1 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 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.

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

An adversarial patch can arbitrarily manipulate image pixels within a restricted region to induce model misclassification. The threat of this localized attack has gained significant attention because the adversary can mount a physicallyrealizable attack by attaching patches to the victim object. Recent provably robust defenses generally follow the PatchGuard framework (Xiang et al., 2021) by using CNNs with small receptive fields and secure feature aggregation for robust model predictions. In this paper, we extend PatchGuard to PatchGuard++ for provably detecting the adversarial patch attack to boost both provable robust accuracy and clean accuracy. In PatchGuard++, we first use a CNN with small receptive fields for feature extraction so that the number of features corrupted by the adversarial patch is bounded. Next, we apply masks in the feature space and evaluate predictions on all possible masked feature maps. Finally, we extract a pattern from all masked predictions to catch the adversarial patch attack. We evaluate PatchGuard++ on ImageNette (a 10-class subset of ImageNet), ImageNet, and CIFAR-10 and demonstrate that PatchGuard++ significantly improves the provable robustness and clean performance.

For agents

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

get_harvested_code_for_paper("2104.12609")
get_code_for_paper("2104.12609")
have("2104.12609")

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

curl https://syntology.ai/api/ran/2104.12609.json

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

[![Syntology run record](https://syntology.ai/api/ran/2104.12609.svg)](https://syntology.ai/paper/2104.12609)

Connect an agent — have() is free.