SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2207.12263 · ECCV · 2022

SecretGen: Privacy Recovery on Pre-trained Models via Distribution Discrimination

Bo Li, Fan Wu, Chaowei Xiao, Zhuowen Yuan, Yunhui Long

arXiv · PDF · Open in the Atlas

Code that ran

8 of 8 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; 1 honoured the contract we drafted, 0 violated it and 0 have no contract verdict; 7 failures of our instrument; 0 no recorded run); 2 of them were also run on the host, not in the jail, by the peer-agreement lane that set their level, 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/2207.12263.json · badge: /api/ran/2207.12263.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
AI-secure/SecretGen canonical (classifier label)
classifier basis not recorded on the link
8 ran · 0 no recorded run
FunctionStatusWhere it lives
np2th Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
AI-secure/SecretGen/tgt_models/vit.py
code served (permissive licence) · get_code("7d8eaf5537f699e4")
How it ran

Ran in Syntology's jail on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1) and returned Tensor, ndim 4, dtype family float, shape (5, 4, 2, 3) as a later run recorded it. The image it ran in was not recorded. The comparison behind this sample's level is a different run: the peer-agreement lane ran it again on the host, not in the jail (report dated 2026-09-05T00:36:31Z). 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 torch
import numpy as np

torch.manual_seed(0)

# Create a small numpy array with shape HWIO (Height, Width, Input channels, Output channels)
# Using distinct dimensions: H=2, W=3, I=4, O=5
weights = np.random.randn(2, 3, 4, 5).astype(np.float32)

ARGS = {
    "weights": weights,
    "conv": True
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 4, dtype family float, shape (5, 4, 2, 3) — shape (5, 4, 2, 3) 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
swish Ran
ran; our contract draft was wrong, not the code
AI-secure/SecretGen/tgt_models/vit.py
code served (permissive licence) · get_code("0f786c407fb1ee4c")
How it ran

Ran in Syntology's jail on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1) and returned Tensor, ndim 2, dtype family float, shape (2, 4) as a later run recorded it. The image it ran in was not recorded. The comparison behind this sample's level is a different run: the peer-agreement lane ran it again on the host, not in the jail (report dated 2026-09-04T22:33:52Z). 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 torch

torch.manual_seed(0)

# The function swish takes a single parameter 'x' which is used with:
# - torch.sigmoid(x) - requires a tensor
# - x * ... - requires a tensor that supports multiplication
# So x should be a tensor

x = torch.randn(2, 4)

ARGS = {
    "x": x
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
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, shape (2, 4) — shape (2, 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
completion_network_loss Ran
ran; our contract draft was wrong, not the code
AI-secure/SecretGen/losses.py
pointer only (licence: MIT); the stored copy get_code serves, from sutd-visual-computing-group/Re-thinking_MI, is pointer only (licence: NONE) · get_code("8ceea87fe22a6f64")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from sutd-visual-computing-group/Re-thinking_MI, and GitHub detected no licence in that repository when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

dconv_bn_relu Ran
ran; our contract draft was wrong, not the code
AI-secure/SecretGen/models.py
pointer only (licence: MIT); the stored copy get_code serves, from ffhibnese/model-inversion-attack-toolbox, is pointer only (licence: NONE) · get_code("5ef80f92b298862d")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from ffhibnese/model-inversion-attack-toolbox, and GitHub detected no licence in that repository when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

get_block Ran
ran, honoured the contract we drafted
AI-secure/SecretGen/facenet.py
code served (permissive licence) · get_code("ad5dbf57f3ea2633")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Sonnet 4.5, prompt python_fixture_v1) and returned list, ndim 2, dtype family int; 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
from collections import namedtuple

torch.manual_seed(0)

# The function get_block takes in_channel, depth, num_units, and stride
# It returns a list of Bottleneck namedtuples

ARGS = {
    'in_channel': 64,
    'depth': 128,
    'num_units': 3,
    'stride': 2
}
Setup block drafted by
a language model, Claude Sonnet 4.5, prompt python_fixture_v1
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type list, ndim 2, dtype family int — 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
get_blocks Ran
ran; our contract draft was wrong, not the code
AI-secure/SecretGen/facenet.py
pointer only (licence: MIT); the stored copy get_code serves, from VladAndronik/face_swap, is pointer only (licence: NONE) · get_code("ea7941eb1ea8ffb2")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from VladAndronik/face_swap, and GitHub detected no licence in that repository when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

l2_norm Ran
ran; our contract draft was wrong, not the code
AI-secure/SecretGen/facenet.py
code served (permissive licence) · get_code("c54fea429589425d")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Sonnet 4.5, prompt python_fixture_v1) and returned Tensor, ndim 2, dtype family float, shape (2, 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 torch

torch.manual_seed(0)

input_tensor = torch.randn(2, 4)
axis = 1

ARGS = {
    'input': input_tensor,
    'axis': axis
}
Setup block drafted by
a language model, Claude Sonnet 4.5, prompt python_fixture_v1
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, shape (2, 4) — shape (2, 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
noise_loss Ran
ran; our contract draft was wrong, not the code
AI-secure/SecretGen/losses.py
pointer only (licence: MIT); the stored copy get_code serves, from ffhibnese/model-inversion-attack-toolbox, is pointer only (licence: NONE) · get_code("32a2722356ae35ea")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from ffhibnese/model-inversion-attack-toolbox, and GitHub detected no licence in that repository when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is 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, 8 are also present in other papers whose code we hold (7 with at least one paper dated earlier, 1 the same month, 8 later) and 0 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.

FunctionAlso present inEarliest-datedOthers named
swish 206 other papers
113 dated earlier, 1 the same month, 88 later, 4 same year, order not determinable · 44 only as an assembly brick
1408.5882 (2014-08)
Convolutional Neural Networks for Sentence Classification
1805.12114 (2018-05)
Deep Reinforcement Learning in a Handful of Trials using Probabilistic Dynamics Models
1809.01357 (2018-09)
Zero Shot Learning for Code Education: Rubric Sampling with Deep Learning Inference
and 203 more not named here
np2th 35 other papers
19 dated earlier, 16 later · 5 only as an assembly brick
2010.11929 (2020-10)
An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale
2102.04306 (2021-02)
TransUNet: Transformers Make Strong Encoders for Medical Image Segmentation
2103.07976 (2021-03)
TransFG: A Transformer Architecture for Fine-grained Recognition
and 32 more not named here
l2_norm 29 other papers
16 dated earlier, 12 later, 1 same year, order not determinable · 2 only as an assembly brick
1804.07573 (2018-04)
MobileFaceNets: Efficient CNNs for Accurate Real-Time Face Verification on Mobile Devices
1903.08527 (2019-03)
Accurate 3D Face Reconstruction with Weakly-Supervised Learning: From Single Image to Image Set
1905.11946 (2019-05)
EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks
and 26 more not named here
get_block 27 other papers
14 dated earlier, 13 later
1804.07573 (2018-04)
MobileFaceNets: Efficient CNNs for Accurate Real-Time Face Verification on Mobile Devices
1903.08527 (2019-03)
Accurate 3D Face Reconstruction with Weakly-Supervised Learning: From Single Image to Image Set
1907.10830 (2019-07)
U-GAT-IT: Unsupervised Generative Attentional Networks with Adaptive Layer-Instance Normalization for Image-to-Image Translation
and 24 more not named here
get_blocks 16 other papers
10 dated earlier, 6 later
1804.07573 (2018-04)
MobileFaceNets: Efficient CNNs for Accurate Real-Time Face Verification on Mobile Devices
1903.08527 (2019-03)
Accurate 3D Face Reconstruction with Weakly-Supervised Learning: From Single Image to Image Set
1907.10830 (2019-07)
U-GAT-IT: Unsupervised Generative Attentional Networks with Adaptive Layer-Instance Normalization for Image-to-Image Translation
and 13 more not named here
completion_network_loss 4 other papers
1 dated earlier, 3 later
2010.04092 (2020-10)
Knowledge-Enriched Distributional Model Inversion Attacks
2402.04013 (2024-02)
Privacy Leakage on DNNs: A Survey of Model Inversion Attacks and Defenses
2411.08460 (2024-11)
Trap-MID: Trapdoor-based Defense against Model Inversion Attacks
and 1 more not named here
dconv_bn_relu 4 other papers
2 dated earlier, 2 later
2010.04092 (2020-10)
Knowledge-Enriched Distributional Model Inversion Attacks
2201.12179 (2022-01)
Plug & Play Attacks: Towards Robust and Flexible Model Inversion Attacks
2402.04013 (2024-02)
Privacy Leakage on DNNs: A Survey of Model Inversion Attacks and Defenses
and 1 more not named here
noise_loss 1 other paper
1 later
2402.04013 (2024-02)
Privacy Leakage on DNNs: A Survey of Model Inversion Attacks and Defenses

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

Transfer learning through the use of pre-trained models has become a growing trend for the machine learning community. Consequently, numerous pre-trained models are released online to facilitate further research. However, it raises extensive concerns on whether these pretrained models would leak privacy-sensitive information of their training data. Thus, in this work, we aim to answer the following questions: "Can we effectively recover private information from these pre-trained models? What are the sufficient conditions to retrieve such sensitive information?" We first explore different statistical information which can discriminate the private training distribution from other distributions. Based on our observations, we propose a novel private data reconstruction framework, SecretGen, to effectively recover private information. Compared with previous methods which can recover private data with the ground truth label of the targeted recovery instance, SecretGen does not require such prior knowledge, making it more practical. We conduct extensive experiments on different datasets under diverse scenarios to compare Secret-Gen with other baselines and provide a systematic benchmark to better understand the impact of different auxiliary information and optimization operations. We show that without prior knowledge about true class prediction, SecretGen is able to recover private data with similar performance compared with the ones that leverage such prior knowledge. If the prior knowledge is given, SecretGen will significantly outperform baseline methods. We also propose several quantitative metrics to further quantify the privacy vulnerability of pre-trained models, which will help the model selection for privacy-sensitive applications. Our code is available at: https://github.com/AI-secure/SecretGen.

For agents

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

get_harvested_code_for_paper("2207.12263")
get_code_for_paper("2207.12263")
have("2207.12263")

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

curl https://syntology.ai/api/ran/2207.12263.json

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

[![Syntology run record](https://syntology.ai/api/ran/2207.12263.svg)](https://syntology.ai/paper/2207.12263)

Connect an agent — have() is free.