SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2605.14686 · 2026

ReMIA: a Powerful and Efficient Alternative to Membership Inference Attacks against Synthetic Data Generators

Andrea Coser, Sebastiano Saccani, Davide Scassola

arXiv · PDF · Open in the Atlas

Code that ran

5 of 8 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (4 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 4 have no contract verdict; 1 failure of our instrument; 3 no recorded run), measured 2026-11-22; 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/2605.14686.json · badge: /api/ran/2605.14686.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
aindo-com/remia community
Papers with Code archive
5 ran · 3 no recorded run
FunctionStatusWhere it lives
DF2Tensor Ran
ran on a synthesized input · constructed an object; no result was computed
aindo-com/remia/src/remia.py
code served (permissive licence) · get_code("cfb068e471b6943b")
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, DF2Tensor: 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 pandas as pd
import torch
import numpy as np
from sklearn.preprocessing import OrdinalEncoder

torch.manual_seed(0)
np.random.seed(0)

# Create a sample dataframe with both numerical and categorical columns
df = pd.DataFrame({
    'num1': [1.0, 2.0],
    'num2': [3.0, 4.0],
    'cat1': ['a', 'b'],
    'cat2': ['x', 'y']
})

ARGS = {
    'one_hot': True,
    'normalize': True
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, pandas, sklearn, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type DF2Tensor — 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 DF2Tensor 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
MLP Ran
ran on a synthesized input · constructed an object; no result was computed
aindo-com/remia/src/remia.py
code served (permissive licence) · get_code("fb579c2299249052")
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, MLP: 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)

# Define ARGS dict with all parameters for MLP.__init__
ARGS = {
    'input_shape': 4,
    'units': [8, 6],
    'dropout': 0.1,
    'activation': nn.ReLU()
}
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 MLP — 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 MLP 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
TabularMLP Ran
ran on a synthesized input · constructed an object; no result was computed
aindo-com/remia/src/remia.py
code served (permissive licence) · get_code("57c535c6037b1856")
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, TabularMLP: 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 numpy as np
import pandas as pd
from sklearn.preprocessing import OrdinalEncoder
import torch.nn as nn

torch.manual_seed(0)

# Create a small sample dataframe with mixed types
df = pd.DataFrame({
    'numeric_1': [1.0, 2.0],
    'numeric_2': [3.0, 4.0],
    'cat_1': ['a', 'b'],
    'cat_2': ['x', 'y'],
    'target': [0, 1]
})

# Define arguments for TabularMLP.__init__
ARGS = {
    'units': [8, 4],  # Hidden layer sizes
    'dropout': 0.1,
    'weight_decay': 0.01,
    'activation': nn.ReLU()
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, pandas, sklearn, torch, tqdm — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type TabularMLP — 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 TabularMLP 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
Timer Ran
ran on a synthesized input · constructed an object; no result was computed
aindo-com/remia/src/remia.py
code served (permissive licence) · get_code("05e023d4fe9438b8")
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, Timer: 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 time

class Timer:

    def __enter__(self):
        self.start = time.time()
        return self

    def __exit__(self, *args):
        self.elapsed = time.time() - self.start


# Timer is a context manager with no __init__ parameters
# It uses __enter__ and __exit__ which are called automatically
# No arguments are needed to construct or use Timer

ARGS = {}
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 Timer — 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 Timer 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
pop_data Ran
ran; our contract draft was wrong, not the code
aindo-com/remia/src/remia.py
code served (permissive licence) · get_code("1cfcba52b8739d7d")
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 DataFrame. 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 pandas as pd

# Create a sample DataFrame with distinct dimensions for probing
df = pd.DataFrame({
    'col_a': [1, 2, 3, 4, 5],
    'col_b': ['x', 'y', 'z', 'w', 'v'],
    'col_c': [10.5, 20.3, 30.1, 40.7, 50.2]
})

n = 2
seed = 42

ARGS = {
    'df': df,
    'n': n,
    'seed': seed
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
pandas — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type DataFrame — a DataFrame output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a DataFrame output the recorded ndim 2 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
Generator Not yet run
unverified (harvested, no recorded run)
aindo-com/remia/src/remia.py
code served (permissive licence) · get_code("8ee797aa2364f1e4")
How it ran

no invocation recorded.

ValidationHook Not yet run
unverified (harvested, no recorded run)
aindo-com/remia/src/remia.py
code served (permissive licence) · get_code("1565280afb298589")
How it ran

no invocation recorded.

remia Not yet run
unverified (harvested, no recorded run)
aindo-com/remia/src/remia.py
code served (permissive licence) · get_code("80830b457bee7b91")
How it ran

no invocation recorded.

Also present in other papers' code

Of 8 samples lifted from this paper's linked repositories, 8 carry a function identity and 0 do not. Of the 8 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 8 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

Tabular data sharing under privacy constraints is increasingly important for research and collaboration. Synthetic data generators (SDGs) are a promising solution, but synthetic data remains vulnerable to attacks, such as membership inference attacks (MIAs), which aim to determine whether a specific record was part of the training data. State-of-the-art MIAs are powerful but impractical: they rely on shadow modeling, requiring hundreds of SDG training runs, and need auxiliary data several times larger than the original training set. Fast proxy metrics like distance to closest record (DCR) are efficient but have limited sensitivity to MIA risk. We introduce ReMIA (Relative Membership Inference Attack), a practical privacy metric that requires only two SDG training runs and additional data no larger than the original training set. Rather than predicting whether a record was in the training set, ReMIA generates two synthetic datasets from two source datasets and measures whether a classifier can identify which source a record came from. Experiments across multiple tabular datasets and SDGs show that ReMIA has a sensitivity comparable to state-of-the-art MIAs while being substantially more practical. We further observe that SDGs can achieve privacy-utility trade-offs that traditional noise-based anonymization methods do not match. Code is available at https://github.com/aindo-com/remia.

For agents

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

get_harvested_code_for_paper("2605.14686")
get_code_for_paper("2605.14686")
have("2605.14686")

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

curl https://syntology.ai/api/ran/2605.14686.json

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

[![Syntology run record](https://syntology.ai/api/ran/2605.14686.svg)](https://syntology.ai/paper/2605.14686)

Connect an agent — have() is free.