SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2501.00824 · 2025

How Breakable Is Privacy: Probing and Resisting Model Inversion Attacks in Collaborative Inference

arXiv · PDF · Open in the Atlas

Code that ran

1 of 4 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; 3 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/2501.00824.json · badge: /api/ran/2501.00824.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
SiftFunnel/SiftFunnel canonical (classifier label)
classifier basis: classifier_2025
1 ran · 3 no recorded run
FunctionStatusWhere it lives
get_pth_paths Ran
ran; our contract draft was wrong, not the code
SiftFunnel/SiftFunnel/train_inversion.py
code served (permissive licence) · get_code("4442ab29464bbc0e")
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 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 re
import tempfile
import shutil

# Create a temporary directory structure with test .pth files
temp_dir = tempfile.mkdtemp()

# Create subdirectories and .pth files matching the expected pattern
os.makedirs(os.path.join(temp_dir, "subdir1"), exist_ok=True)
os.makedirs(os.path.join(temp_dir, "subdir2"), exist_ok=True)

# Create test .pth files with names matching the pattern: {dataset}_{target_model}.pth
test_file_1 = os.path.join(temp_dir, "subdir1", "cifar10_resnet50.pth")
test_file_2 = os.path.join(temp_dir, "subdir2", "cifar10_resnet18_pretrained.pth")
test_file_3 = os.path.join(temp_dir, "cifar10_vgg16.pth")

# Create empty .pth files
open(test_file_1, 'w').close()
open(test_file_2, 'w').close()
open(test_file_3, 'w').close()

# Also create some non-matching files to test filtering
open(os.path.join(temp_dir, "imagenet_resnet50.pth"), 'w').close()
open(os.path.join(temp_dir, "other_file.txt"), 'w').close()

ARGS = {
    "base_dir": temp_dir,
    "dataset": "cifar10",
    "target_model": "resnet"
}
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 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
cosine_similarity Not yet run
unverified (harvested, no recorded run)
SiftFunnel/SiftFunnel/train_inversion.py
code served (permissive licence) · get_code("902096bb1a33d741")
How it ran

no invocation recorded.

mine_loss Not yet run
unverified (harvested, no recorded run)
SiftFunnel/SiftFunnel/Mutual_Information_estimator.py
code served (permissive licence) · get_code("8c95fdfb2c19483c")
How it ran

no invocation recorded.

psnr Not yet run
unverified (harvested, no recorded run)
SiftFunnel/SiftFunnel/train_inversion.py
code served (permissive licence) · get_code("133fad3eb5762282")
How it ran

no invocation recorded.

Also present in other papers' code

Of 4 samples lifted from this paper's linked repositories, 4 carry a function identity and 0 do not. Of the 4 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 4 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

Collaborative inference (CI) improves computational efficiency for edge devices by transmitting intermediate features to cloud models. However, this process inevitably exposes feature representations to model inversion attacks (MIAs), enabling unauthorized data reconstruction. Despite extensive research, there is no established criterion for assessing the difficulty of MIA implementation, leaving a fundamental question unanswered: \textit{What factors truly and verifiably determine the attack's success in CI?} Moreover, existing defenses lack the theoretical foundation described above, making it challenging to regulate feature information effectively while ensuring privacy and minimizing computational overhead. These shortcomings introduce three key challenges: theoretical gap, methodological limitation, and practical constraint. To overcome these challenges, we propose the first theoretical criterion to assess MIA difficulty in CI, identifying mutual information, entropy, and effective information volume as key influencing factors. The validity of this criterion is demonstrated by using the mutual information neural estimator. Building on this insight, we propose SiftFunnel, a privacy-preserving framework to resist MIA while maintaining usability. Specifically, we incorporate linear and non-linear correlation constraints alongside label smoothing to suppress redundant information transmission, effectively balancing privacy and usability. To enhance deployability, the edge model adopts a funnel-shaped structure with attention mechanisms, strengthening privacy while reducing computational and storage burdens. Experiments show that, compared to state-of-the-art defense, SiftFunnel increases reconstruction error by $\sim$30\%, lowers mutual and effective information metrics by $\geq$50\%, and reduces edge burdens by almost $20\times$, while maintaining comparable usability.

For agents

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

get_harvested_code_for_paper("2501.00824")
get_code_for_paper("2501.00824")
have("2501.00824")

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

curl https://syntology.ai/api/ran/2501.00824.json

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

[![Syntology run record](https://syntology.ai/api/ran/2501.00824.svg)](https://syntology.ai/paper/2501.00824)

Connect an agent — have() is free.