SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2603.29410 · 2026

AGFT: Alignment-Guided Fine-Tuning for Zero-Shot Adversarial Robustness of Vision-Language Models

Zheng Zhang, Yubo Cui, Xianchao Guan, Zijun Xiong

arXiv · PDF · Open in the Atlas

Code that ran

4 of 6 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; 4 failures of our instrument; 2 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/2603.29410.json · badge: /api/ran/2603.29410.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
YuboCui/AGFT canonical (classifier label)
classifier basis: census
4 ran · 2 no recorded run
FunctionStatusWhere it lives
get_train_dataset Ran
ran; our contract draft was wrong, not the code
YuboCui/AGFT/utils.py
pointer only (licence: NONE) · get_code("cd82db2984e88662")
How it ran

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

load_folder2name Ran
ran; our contract draft was wrong, not the code
YuboCui/AGFT/clip_dataset.py
pointer only (licence: NONE) · get_code("523d7c2de2a69ace")
How it ran

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

refine_classname Ran
ran; our contract draft was wrong, not the code
YuboCui/AGFT/clip_dataset.py
pointer only (licence: NONE) · get_code("801704a2c8af559e")
How it ran

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

validate_one Ran
ran; our contract draft was wrong, not the code
YuboCui/AGFT/validate.py
pointer only (licence: NONE) · get_code("e34f1b1db45c292d")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, YuboCui/AGFT 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_val_attack Not yet run
unverified (harvested, no recorded run)
YuboCui/AGFT/utils.py
pointer only (licence: NONE) · get_code("3a72468987fa3206")
How it ran

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

get_val_dataset Not yet run
unverified (harvested, no recorded run)
YuboCui/AGFT/utils.py
pointer only (licence: NONE) · get_code("5ed58d803dcaec0f")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, YuboCui/AGFT 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 6 samples lifted from this paper's linked repositories, 6 carry a function identity and 0 do not. Of the 6 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 6 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

Pre-trained vision-language models (VLMs) exhibit strong zero-shot generalization but remain vulnerable to adversarial perturbations. Existing classification-guided adversarial fine-tuning methods often disrupt pre-trained crossmodal alignment, weakening visual-textual correspondence and degrading zero-shot performance. In this paper, we propose an Alignment-Guided Fine-Tuning (AGFT) framework that enhances zero-shot adversarial robustness while preserving the cross-modal semantic structure. Unlike label-based methods that rely on hard labels and fail to maintain the relative relationships between image and text, AGFT leverages the probabilistic predictions of the original model for text-guided adversarial training, which aligns adversarial visual features with textual embeddings via soft alignment distributions, improving zero-shot adversarial robustness. To address structural discrepancies introduced by fine-tuning, we introduce a distribution consistency calibration mechanism that adjusts the robust model's output to match a temperature-scaled version of the pre-trained model's predictions. Extensive experiments across multiple zero-shot benchmarks demonstrate that AGFT outperforms state-of-the-art methods while significantly improving zero-shot adversarial robustness. Code is available at https://github.com/YuboCui/AGFT.

For agents

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

get_harvested_code_for_paper("2603.29410")
get_code_for_paper("2603.29410")
have("2603.29410")

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

curl https://syntology.ai/api/ran/2603.29410.json

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

[![Syntology run record](https://syntology.ai/api/ran/2603.29410.svg)](https://syntology.ai/paper/2603.29410)

Connect an agent — have() is free.