SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1903.02709 · 2019

On Adversarial Mixup Resynthesis

arXiv · PDF · Open in the Atlas

Code that ran

6 of 7 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 6 have no contract verdict; 0 failures of our instrument; 1 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/1903.02709.json · badge: /api/ran/1903.02709.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
christopher-beckham/amr canonical (classifier label)
classifier basis: census
6 ran · 1 no recorded run
FunctionStatusWhere it lives
find_latest_pkl_in_folder Ran
ran on a synthesized input
christopher-beckham/amr/tools.py
code served (permissive licence) · get_code("5552de7ae2b53587")
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 str. 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 os
import tempfile
import glob

# Create a temporary directory with some .pkl files
temp_dir = tempfile.mkdtemp()

# Create some dummy .pkl files with different modification times
pkl_file_1 = os.path.join(temp_dir, "model_1.pkl")
pkl_file_2 = os.path.join(temp_dir, "model_2.pkl")
pkl_file_3 = os.path.join(temp_dir, "model_3.pkl")

# Create the files
with open(pkl_file_1, 'w') as f:
    f.write("dummy")

with open(pkl_file_2, 'w') as f:
    f.write("dummy")

with open(pkl_file_3, 'w') as f:
    f.write("dummy")

# Set different modification times so we can test max(files, key=os.path.getctime)
os.utime(pkl_file_1, (1000000000, 1000000000))
os.utime(pkl_file_2, (1000000100, 1000000100))
os.utime(pkl_file_3, (1000000050, 1000000050))

ARGS = {
    'model_dir': temp_dir
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
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 str — a str output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a str 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
generate_name_from_args Ran
ran on a synthesized input
christopher-beckham/amr/tools.py
code served (permissive licence) · get_code("5751304a8a67f903")
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 str. 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 numpy as np
from collections import OrderedDict

torch.manual_seed(0)

# Construct the arguments for generate_name_from_args
dd = {
    'learning_rate': 0.001,
    'batch_size': 32,
    'dropout': 0.5,
    'epochs': 10,
    'use_cuda': True,
    'model_name': None,
    'hidden_dim': 128
}

kwargs_for_name = {
    'learning_rate': ('lr', lambda x: f'{x:.4f}'),
    'batch_size': ('bs', lambda x: str(x)),
    'dropout': ('drop', lambda x: f'{x:.1f}'),
    'epochs': ('ep', lambda x: str(x)),
    'use_cuda': ('cuda', lambda x: 'yes'),
    'hidden_dim': ('hdim', lambda x: str(x))
}

ARGS = {
    'dd': dd,
    'kwargs_for_name': kwargs_for_name
}
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 str — a str output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a str 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
get_shuffled_indices Ran
ran on a synthesized input
christopher-beckham/amr/task_launcher.py
code served (permissive licence) · get_code("7db736b5986a5c7a")
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 int, shape (10,) 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

def get_shuffled_indices(length):
    rnd_state = np.random.RandomState(0)
    idxs = np.arange(length)
    rnd_state.shuffle(idxs)
    return idxs

ARGS = {
    "length": 10
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
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 1, dtype family int, shape (10,) — shape (10,) 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
image_handler_default Ran
ran on a synthesized input
christopher-beckham/amr/handlers.py
code served (permissive licence) · get_code("b8f6608294d5b9a2")
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 function. 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 tempfile
import os

torch.manual_seed(0)

# Create a temporary directory for saving images
temp_dir = tempfile.mkdtemp()

ARGS = {
    'save_path': temp_dir,
    'save_images_every': 1
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch, torchvision — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type function — a function output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a function 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
line2dict Ran
ran on a synthesized input
christopher-beckham/amr/tools.py
code served (permissive licence) · get_code("4af39f15dcb09084")
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 dict. 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

def line2dict(st):
    """Convert a line of key=value pairs to a
    dictionary.

    :param st:
    :returns: a dictionary
    :rtype:
    """
    elems = st.split(',')
    dd = {}
    for elem in elems:
        elem = elem.split('=')
        key, val = elem
        try:
            int_val = int(val)
            dd[key] = int_val
        except ValueError:
            dd[key] = val
    return dd


ARGS = {
    "st": "learning_rate=0.001,batch_size=32,epochs=10,model=resnet50"
}
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 dict — a dict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a dict 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
test_set_handler Ran
ran on a synthesized input
christopher-beckham/amr/handlers.py
code served (permissive licence) · get_code("a7914621e62896fa")
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 function. 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
import numpy as np
import tempfile

torch.manual_seed(0)

class SimpleGenerator(nn.Module):
    def __init__(self):
        super().__init__()
        self.encoder = nn.Linear(4, 8)
    
    def encode(self, x):
        return self.encoder(x)

class SimpleClassifier(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear = nn.Linear(8, 3)
        self.n_in = 8
    
    def forward(self, x):
        return self.linear(x)

class SimpleGAN(nn.Module):
    def __init__(self):
        super().__init__()
        self.generator = SimpleGenerator()
        self.cls_enc = SimpleClassifier()
        self.use_cuda = False

gan = SimpleGAN()

# Create a simple dataset as list of (x, y) tuples
# x: batch of 2 samples, 4 features
# y: batch of 2 one-hot encoded labels (3 classes)
dataset = [
    (torch.randn(2, 4), torch.tensor([[1, 0, 0], [0, 1, 0]], dtype=torch.float32)),
    (torch.randn(2, 4), torch.tensor([[0, 0, 1], [1, 0, 0]], dtype=torch.float32)),
]

save_path = tempfile.mkdtemp()

ARGS = {
    'gan': gan,
    'dataset': dataset,
    'save_path': save_path,
}
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 function — a function output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a function 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
get_network Not yet run
unverified (harvested, no recorded run)
christopher-beckham/amr/architectures/arch_acai_kyle.py
code served (permissive licence) · get_code("4042965d38c66087")
How it ran

no invocation recorded.

Also present in other papers' code

Of 7 samples lifted from this paper's linked repositories, 7 carry a function identity and 0 do not. Of the 7 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 7 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

In this paper, we explore new approaches to combining information encoded within the learned representations of auto-encoders. We explore models that are capable of combining the attributes of multiple inputs such that a resynthesised output is trained to fool an adversarial discriminator for real versus synthesised data. Furthermore, we explore the use of such an architecture in the context of semi-supervised learning, where we learn a mixing function whose objective is to produce interpolations of hidden states, or masked combinations of latent representations that are consistent with a conditioned class label. We show quantitative and qualitative evidence that such a formulation is an interesting avenue of research.

For agents

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

get_harvested_code_for_paper("1903.02709")
get_code_for_paper("1903.02709")
have("1903.02709")

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

curl https://syntology.ai/api/ran/1903.02709.json

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

[![Syntology run record](https://syntology.ai/api/ran/1903.02709.svg)](https://syntology.ai/paper/1903.02709)

Connect an agent — have() is free.