SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2010.06154 · 2020

An Analysis of Robustness of Non-Lipschitz Networks

arXiv · PDF · Open in the Atlas

Code that ran

3 of 3 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; 2 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/2010.06154.json · badge: /api/ran/2010.06154.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
dravyanshsharma/adversarial-contrastive canonical (classifier label)
classifier basis: classifier_2025
3 ran · 0 no recorded run
FunctionStatusWhere it lives
load_data Ran
ran; our contract draft was wrong, not the code
dravyanshsharma/adversarial-contrastive/n3_1__optimal_attack.py
code served (permissive licence) · get_code("50768e5c6de22234")
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 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 numpy as np
import os
import tempfile

# Create a temporary directory structure with mock data files
temp_dir = tempfile.mkdtemp()
features_dir = os.path.join(temp_dir, 'features')
os.makedirs(features_dir, exist_ok=True)

# Create mock numpy arrays for different type and tau combinations
def create_mock_files(type_str, tau_val=None):
    taustr = ''
    if tau_val:
        taustr = '_' + str(tau_val)
    
    # Create small mock data
    train_features = np.random.randn(10, 5).astype(np.float32)
    train_labels = np.random.randint(0, 2, 10).astype(np.int64)
    test_features = np.random.randn(5, 5).astype(np.float32)
    test_labels = np.random.randint(0, 2, 5).astype(np.int64)
    
    # Save files
    np.save(os.path.join(features_dir, f'{type_str}con_feature_train{taustr}.npy'), train_features)
    np.save(os.path.join(features_dir, f'{type_str}con_label_train{taustr}.npy'), train_labels)
    np.save(os.path.join(features_dir, f'{type_str}con_feature_test.npy'), test_features)
    np.save(os.path.join(features_dir, f'{type_str}con_label_test.npy'), test_labels)

# Create mock files for the test case
create_mock_files('image', tau_val=0.5)
create_mock_files('image', tau_val=None)

# Change to temp directory so relative paths work
original_dir = os.getcwd()
os.chdir(temp_dir)

# Define arguments for the function
ARGS = {
    'type': 'image',
    'tau': 0.5
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
numpy — 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
make_rand_vector Ran
ran, honoured the contract we drafted
dravyanshsharma/adversarial-contrastive/n3_1__optimal_attack.py
code served (permissive licence) · get_code("14668e20fe120e68")
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 list, ndim 1, 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.

from random import gauss

def make_rand_vector(dims):
    vec = [gauss(0, 1) for i in range(dims)]
    mag = sum((x ** 2 for x in vec)) ** 0.5
    return [x / mag for x in vec]

ARGS = {
    "dims": 5
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type list, ndim 1, 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
rotation_matrix Ran
ran, honoured the contract we drafted
dravyanshsharma/adversarial-contrastive/n3_1__optimal_attack.py
code served (permissive licence) · get_code("d936aa6a51d78d60")
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 ndarray, ndim 2, dtype family float, shape (4, 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 numpy as np

# Analyze the function:
# - Takes two parameters: a and b
# - Performs matrix operations: transpose, matmul, addition
# - Uses len(a) to get dimension
# - Returns a square matrix of size len(a) x len(a)
# 
# The function expects:
# - a: a 1D or 2D array-like that supports np.transpose, np.matmul, addition
# - b: same as a
# - Both should have the same shape for addition to work
# - len(a) should give the dimension for np.identity

# Create small numpy arrays as vectors
a = np.array([1.0, 2.0, 3.0, 4.0])
b = np.array([0.5, 1.5, 2.5, 3.5])

ARGS = {
    "a": a,
    "b": b
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
numpy — 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 (4, 4) — shape (4, 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

Also present in other papers' code

Of 3 samples lifted from this paper's linked repositories, 3 carry a function identity and 0 do not. Of the 3 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 3 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

Despite significant advances, deep networks remain highly susceptible to adversarial attack. One fundamental challenge is that small input perturbations can often produce large movements in the network's final-layer feature space. In this paper, we define an attack model that abstracts this challenge, to help understand its intrinsic properties. In our model, the adversary may move data an arbitrary distance in feature space but only in random low-dimensional subspaces. We prove such adversaries can be quite powerful: defeating any algorithm that must classify any input it is given. However, by allowing the algorithm to abstain on unusual inputs, we show such adversaries can be overcome when classes are reasonably well-separated in feature space. We further provide strong theoretical guarantees for setting algorithm parameters to optimize over accuracy-abstention trade-offs using data-driven methods. Our results provide new robustness guarantees for nearest-neighbor style algorithms, and also have application to contrastive learning, where we empirically demonstrate the ability of such algorithms to obtain high robust accuracy with low abstention rates. Our model is also motivated by strategic classification, where entities being classified aim to manipulate their observable features to produce a preferred classification, and we provide new insights into that area as well.

For agents

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

get_harvested_code_for_paper("2010.06154")
get_code_for_paper("2010.06154")
have("2010.06154")

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

curl https://syntology.ai/api/ran/2010.06154.json

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

[![Syntology run record](https://syntology.ai/api/ran/2010.06154.svg)](https://syntology.ai/paper/2010.06154)

Connect an agent — have() is free.