SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2601.12263 · 2026

Multimodal Generative Engine Optimization: Rank Manipulation for Vision-Language Model Rankers

Yue Zhao, Ziyi Wang, Xiyang Hu, Yixuan Du, Chenxiao Yu, Haoyan Xu

arXiv · PDF · Open in the Atlas

Code that ran

13 of 15 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; 5 honoured the contract we drafted, 0 violated it and 0 have no contract verdict; 8 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/2601.12263.json · badge: /api/ran/2601.12263.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
glad-lab/MGEO canonical (classifier label)
classifier basis: census
13 ran · 2 no recorded run
FunctionStatusWhere it lives
bleu_loss Ran
ran; our contract draft was wrong, not the code
glad-lab/MGEO/attack.py
pointer only (licence: NONE) · get_code("38945963615f4620")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, glad-lab/MGEO 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.

extract_brand Ran
ran; our contract draft was wrong, not the code
glad-lab/MGEO/clean_product_names.py
pointer only (licence: NONE) · get_code("33dd6035c1d1f210")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, glad-lab/MGEO 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.

extract_category_from_path Ran
ran; our contract draft was wrong, not the code
glad-lab/MGEO/clean_product_names.py
pointer only (licence: NONE) · get_code("5782194a87c75177")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, glad-lab/MGEO 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.

fluency_loss Ran
ran; our contract draft was wrong, not the code
glad-lab/MGEO/attack.py
pointer only (licence: NONE) · get_code("f251acd017c5ac40")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, glad-lab/MGEO 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_banned_tokens Ran
ran, honoured the contract we drafted
glad-lab/MGEO/attack_autodan.py
pointer only (licence: NONE) · get_code("d35455c8c31a85ad")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, glad-lab/MGEO 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_clean_names_prompt Ran
ran; our contract draft was wrong, not the code
glad-lab/MGEO/clean_product_names.py
pointer only (licence: NONE) · get_code("5ec37768d9f9fde9")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, glad-lab/MGEO 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_embedding_matrix Ran
ran, honoured the contract we drafted
glad-lab/MGEO/attack_autodan.py
pointer only (licence: NONE) · get_code("27171d1b04633a6f")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, glad-lab/MGEO 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_embeddings Ran
ran; our contract draft was wrong, not the code
glad-lab/MGEO/attack_autodan.py
pointer only (licence: NONE) · get_code("19c802b4dd21c8b8")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, glad-lab/MGEO 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_user_query Ran
ran; our contract draft was wrong, not the code
glad-lab/MGEO/attack_config.py
pointer only (licence: NONE) · get_code("02a7330e94439059")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, glad-lab/MGEO 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.

init_prompt Ran
ran, honoured the contract we drafted
glad-lab/MGEO/process.py
pointer only (licence: NONE) · get_code("4c8c8c4eb9978e7e")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, glad-lab/MGEO 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.

process_headtail Ran
ran, honoured the contract we drafted
glad-lab/MGEO/process.py
pointer only (licence: NONE) · get_code("e8b5ab5b2a10f6b5")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, glad-lab/MGEO 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.

process_text Ran
ran, honoured the contract we drafted
glad-lab/MGEO/process.py
pointer only (licence: NONE) · get_code("80e2a549a1ab5ead")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, glad-lab/MGEO 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.

soft_forward Ran
ran; our contract draft was wrong, not the code
glad-lab/MGEO/attack.py
pointer only (licence: NONE) · get_code("6648c4d59ef6b0f4")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, glad-lab/MGEO 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.

generate_adversarial_image_baseline Not yet run
unverified (harvested, no recorded run)
glad-lab/MGEO/baseline.py
pointer only (licence: NONE) · get_code("a2502f1d9c6c1d90")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, glad-lab/MGEO 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.

generate_adversarial_text_baseline Not yet run
unverified (harvested, no recorded run)
glad-lab/MGEO/baseline.py
pointer only (licence: NONE) · get_code("cfc913dec684f279")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, glad-lab/MGEO 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 15 samples lifted from this paper's linked repositories, 15 carry a function identity and 0 do not. Of the 15 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 15 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

Vision-Language Models (VLMs) integrate visual and textual knowledge into unified representations that increasingly underpin modern retrieval and recommendation systems. However, it remains unclear how reliably these models utilize their cross-modal knowledge when ranking multimodal items, and whether their knowledge grounding can be subverted. In this paper, we expose a fundamental vulnerability in how VLMs apply multimodal knowledge for product ranking: through Multimodal Generative Engine Optimization (MGEO), we show that an adversary can manipulate a VLM's ranking decisions by jointly crafting imperceptible image perturbations and fluent textual suffixes that exploit the model's internal cross-modal knowledge coupling. Using an alternating optimization strategy, MGEO targets the deep interactions between visual and linguistic representations within the VLM, achieving rank manipulations that substantially exceed those of unimodal attacks and heuristic baselines powered by strong commercial models. Our findings reveal that surface-level content quality is insufficient for rank promotion; instead, direct alignment with the model's internal knowledge utilization mechanism is required. These results raise important questions on the faithfulness and robustness of knowledge grounding in multimodal foundation models, and motivate future work on defense mechanisms for multimodal retrieval systems. 1 * Corresponding author. 1 Code and data: https://github.com/glad-lab/MGEO Recent work has exposed the fragility of Large Language Model (LLM)-based ranking pipelines to textual adversarial manipulation (Hu, 2025). A growing line of research demonstrates that malicious actors can inject carefully crafted textual triggers into website contents or product descriptions to hijack search rankings without altering the underlying model. Techniques range from rewriting

For agents

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

get_harvested_code_for_paper("2601.12263")
get_code_for_paper("2601.12263")
have("2601.12263")

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

curl https://syntology.ai/api/ran/2601.12263.json

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

[![Syntology run record](https://syntology.ai/api/ran/2601.12263.svg)](https://syntology.ai/paper/2601.12263)

Connect an agent — have() is free.