SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2502.18508 · ICLR · 2025

REFINE: Inversion-Free Backdoor Defense via Model Reprogramming

Pin-Yu Chen, Yiming Li, Yukun Chen, Kui Ren, Zhan Qin, Shuo Shao, Enhao Huang

arXiv · PDF · Open in the Atlas

Code that ran

5 of 10 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (1 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; 5 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/2502.18508.json · badge: /api/ran/2502.18508.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
whitolfchen/refine canonical (classifier label)
classifier basis: classifier_2025
3 ran · 4 no recorded run
thuyimingli/backdoorbox canonical (classifier label)
classifier basis: classifier_2025
2 ran · 1 no recorded run
FunctionStatusWhere it lives
Base Ran
ran on a synthesized input · constructed an object; no result was computed
thuyimingli/backdoorbox/core/defenses/REFINE.py
pointer only (licence: GPL-2.0) · get_code("ec367d1aa2f8a7e3")
How it ran

The body of this sample is withheld: GPL-2.0, as GitHub detects it on this paper's repository, thuyimingli/backdoorbox, is not in the permissive set Syntology serves inline. 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.

GetDataset Ran
ran on a synthesized input
whitolfchen/refine/refine.py
pointer only (licence: NONE) · get_code("fdba41a2fab1727d")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, whitolfchen/refine 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.

Log Ran
ran on a synthesized input
whitolfchen/refine/refine.py
pointer only (licence: NONE) · get_code("ec3a04dbcd72b9ff")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, whitolfchen/refine 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.

UNetLittle Ran
ran on a synthesized input
whitolfchen/refine/refine.py
pointer only (licence: NONE) · get_code("d77a5255fcd4aac7")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, whitolfchen/refine 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.

accuracy Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
thuyimingli/backdoorbox/core/defenses/REFINE.py
pointer only (licence: GPL-2.0) · get_code("c70af6aa43a490cb")
How it ran

The body of this sample is withheld: GPL-2.0, as GitHub detects it on this paper's repository, thuyimingli/backdoorbox, is not in the permissive set Syntology serves inline. 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.

Adv_Program Not yet run
unverified (harvested, no recorded run)
whitolfchen/refine/refine.py
pointer only (licence: NONE) · get_code("a6d28cbceecb86a7")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, whitolfchen/refine 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. No invocation recorded.

Adversarial_Reprogramming Not yet run
unverified (harvested, no recorded run)
whitolfchen/refine/refine.py
pointer only (licence: NONE) · get_code("7b1ebd0e995b56df")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, whitolfchen/refine 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. No invocation recorded.

GetModel Not yet run
unverified (harvested, no recorded run)
whitolfchen/refine/refine.py
pointer only (licence: NONE) · get_code("0741157e75879495")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, whitolfchen/refine 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. No invocation recorded.

REFINE Not yet run
unverified (harvested, no recorded run)
thuyimingli/backdoorbox/core/defenses/REFINE.py
pointer only (licence: GPL-2.0) · get_code("0a1de7f4e9eb05db")
How it ran

The body of this sample is withheld: GPL-2.0, as GitHub detects it on this paper's repository, thuyimingli/backdoorbox, is not in the permissive set Syntology serves inline. 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. No invocation recorded.

Up Not yet run
unverified (harvested, no recorded run)
whitolfchen/refine/refine.py
pointer only (licence: NONE) · get_code("090400a79032c4ed")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, whitolfchen/refine 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. No invocation recorded.

Also present in other papers' code

Of 10 samples lifted from this paper's linked repositories, 10 carry a function identity and 0 do not. Of the 10 distinct functions with an identity, 3 are also present in other papers whose code we hold (3 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.

FunctionAlso present inEarliest-datedOthers named
accuracy
brick of this paper's assembly: metric
7 other papers
7 dated earlier
1806.09055 (2018-06)
DARTS: Differentiable Architecture Search
2103.15798 (2021-03)
Rethinking Neural Operations for Diverse Tasks
2109.03508 (2021-09)
RepNAS: Searching for Efficient Re-parameterizing Blocks
and 4 more not named here
Up
brick of this paper's assembly: helper
3 other papers
3 dated earlier
1505.04597 (2015-05)
U-Net: Convolutional Networks for Biomedical Image Segmentation
2011.05479 (2020-11)
ForestNet: Classifying Drivers of Deforestation in Indonesia using Deep Learning on Satellite Imagery
2309.04814 (2023-09)
Speech2Lip: High-fidelity Speech to Lip Generation by Learning from a Short Video
Base
brick of this paper's assembly: helper
1 other paper
1 dated earlier
2405.09786 (2024-05)
IBD-PSC: Input-level Backdoor Detection via Parameter-oriented Scaling Consistency

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

Backdoor attacks on deep neural networks (DNNs) have emerged as a significant security threat, allowing adversaries to implant hidden malicious behaviors during the model training phase. Pre-processing-based defense, which is one of the most important defense paradigms, typically focuses on input transformations or backdoor trigger inversion (BTI) to deactivate or eliminate embedded backdoor triggers during the inference process. However, these methods suffer from inherent limitations: transformation-based defenses often fail to balance model utility and defense performance, while BTI-based defenses struggle to accurately reconstruct trigger patterns without prior knowledge. In this paper, we propose REFINE, an inversion-free backdoor defense method based on model reprogramming. RE-FINE consists of two key components: (1) an input transformation module that disrupts both benign and backdoor patterns, generating new benign features; and (2) an output remapping module that redefines the model's output domain to guide the input transformations effectively. By further integrating supervised contrastive loss, REFINE enhances the defense capabilities while maintaining model utility. Extensive experiments on various benchmark datasets demonstrate the effectiveness of our REFINE and its resistance to potential adaptive attacks.

For agents

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

get_harvested_code_for_paper("2502.18508")
get_code_for_paper("2502.18508")
have("2502.18508")

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

curl https://syntology.ai/api/ran/2502.18508.json

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

[![Syntology run record](https://syntology.ai/api/ran/2502.18508.svg)](https://syntology.ai/paper/2502.18508)

Connect an agent — have() is free.