Papers › VisCRA: A Visual Chain Reasoning Attack for Jailbreaking Multimodal Large Language Models

VisCRA: A Visual Chain Reasoning Attack for Jailbreaking Multimodal Large Language Models

26 May 2025arXiv:2505.19684archive 2025-07-28

Bingrui Sima, Linhua Cong, Wenxuan Wang, Kun He

The emergence of Multimodal Large Language Models (MLRMs) has enabled sophisticated visual reasoning capabilities by integrating reinforcement learning and Chain-of-Thought (CoT) supervision. However, while these enhanced reasoning capabilities improve performance, they also introduce new and underexplored safety risks. In this work, we systematically investigate the security implications of advanced visual reasoning in MLRMs. Our analysis reveals a fundamental trade-off: as visual reasoning improves, models become more vulnerable to jailbreak attacks. Motivated by this critical finding, we introduce VisCRA (Visual Chain Reasoning Attack), a novel jailbreak framework that exploits the visual reasoning chains to bypass safety mechanisms. VisCRA combines targeted visual attention masking with a two-stage reasoning induction strategy to precisely control harmful outputs. Extensive experiments demonstrate VisCRA's significant effectiveness, achieving high attack success rates on leading closed-source MLRMs: 76.48% on Gemini 2.0 Flash Thinking, 68.56% on QvQ-Max, and 56.60% on GPT-4o. Our findings highlight a critical insight: the very capability that empowers MLRMs -- their visual reasoning -- can also serve as an attack vector, posing significant security risks.

PaperPDFCode Syntology ran

In Syntology View this paper on Syntology, its page in Syntology's graph. That page adds which of its functions are also present in other papers' code Syntology holds, with dates, and its dated run record as JSON and a README badge. It also lists the repositories linked to the paper, the abstract and the calls for agents.

Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

For agents, on Syntology's MCP service (how to connect):

Code

Syntology 12 code samples harvested from 1 repository linked to this paper: 9 ran (of which 0 constructed an object rather than computing a result; 7 with no instrument failure: 0 honoured, 0 violated, 7 with no contract checked; 2 where Syntology's instrument failed) · 3 unverified. Where Syntology's instrument failed: 2 where our contract draft was wrong; failures of Syntology's instrument, not of the code.

By repository: found in paper text by Syntology: 12 samples from 1 repository, 9 ran (7 with no instrument failure; 2 where Syntology's instrument failed). The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

Levels written between 2026-09-22 and 2026-09-25 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed.

DyMessi/VisCRA found in paper text by SyntologyApache-2.0Syntology: 12 samples, 9 ran (2 instrument-failed) · 3 unverified report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

12 samples harvested: 9 ran (of which 0 constructed an object rather than computing a result; 7 with no instrument failure: 0 honoured, 0 violated, 7 with no contract checked; 2 where Syntology's instrument failed) · 3 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page last changed ; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

2ran · our draft was wrong
7ran
3unverified

Licence: 2 of the 12 samples are pointer only, meaning Syntology does not serve that copy's text. For 2 of them, this paper's copy is under a permissive licence and cleared, but the stored copy of the same code that Syntology serves is not; the licence label's tooltip gives Syntology's recorded reason. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from DyMessi/VisCRA. “Ran” means the sample executed on a synthesized input. It does not mean the output is correct, and nothing here reproduces the paper's results. A line beginning “fixture:” under a sample that ran gives the arguments Syntology's fixture called the sample with, each as name:type/rank/element kind (? where none was recorded), and the type of what the fixture's run returned, exactly as the harness recorded them; the arguments and the returned type were recorded by two separate runs of the same fixture. No output value is shown, and a returned type says nothing about whether the output was right. “Level recorded” gives the UTC date when Syntology's graph wrote the sample's verification level, not when the code executed; where the level was written before Syntology's graph kept that date, the row says “level recorded, date not recorded”. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself. “Our draft was wrong” and “the contract check could not run” are failures of Syntology's instrument, not of the code: the sample ran on Syntology's synthesized input, and then the drafted contract did not describe it, or the check itself raised an error (RAISES), lacked a dependency (DEP_MISSING) or ran out of time (TIMEOUT); hover the label for which. A metamorphic tier, where shown, is evidence of the output's shape, not that the sample implements the method.

Each sample ends with its code_sha256. A sample's code_sha256 is Syntology's identity for that exact code. It is printed in full (16 hexadecimal characters for every sample in this build), and get_code(code_sha256="…") takes it exactly as printed. An agent calls it on Syntology's MCP service (how to connect); click an identity to copy that call.

Repository labels, per sample. official repository: The archive marks this repository official for the paper. named in the paper: The archive records that the paper mentions this repository; it is not marked official. community (archive-listed): In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper. found in paper text by Syntology: Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted. community: Not in the archive's code links for this paper; a community repository Syntology harvested. Samples are grouped by repository: a repository marked official first, then by how directly the repository is tied to the paper, then in the archive's order; never by how many ran. Licence labels name the repository's licence as recorded at harvest. “Pointer only” means Syntology does not serve that copy's text, for one of six reasons: no licence file was found; the licence was not identified; the licence is recorded as permissive but that copy's record is not marked cleared; the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar); this paper's copy was not located, so its licence is not recorded; or this paper's copy is cleared, but the stored copy of the same code that Syntology serves is not. Some licences outside that list permit redistribution, such as WTFPL, and GPL-3.0 under its conditions; they are simply not on the list. Hover a licence label for the reason. File links open the file on GitHub at the default branch, which may have changed since the harvest.

A function or class name links to its page when the name occurs in the harvested code of at least 5 papers; the name is a string, not an identity.

DyMessi/VisCRA found in paper text by Syntology 12 samples, 9 ran (2 instrument-failed) · 3 unverified
encode_image in 131 papers evaluation/gemini.py ran · our draft was wrong Apache-2.0 (permissive) here; the stored copy is pointer only · f41cb1a19b154297 · report fixture: signature image_path:str/0/str · returned str · level recorded 2026-09-22
encode_image in 131 papers evaluation/QvQ_Max.py ran Apache-2.0 (permissive) · cfe427db2f2f5fcc · report fixture: signature image_path:str/0/str · returned dict · level recorded 2026-09-25
find_top3_blocks_with_stride qwenmask.py ran Apache-2.0 (permissive) · 747918e3fc43514d · report fixture: signature matrix:ndarray/2/float, block_size:int/0/int, stride:int/0/int, max_overlap:int/0/int · returned list · level recorded 2026-09-25
guess_image_type_from_base64 evaluation/gemini.py ran · our draft was wrong fingerprinted Apache-2.0 (permissive) here; the stored copy is pointer only · 8176c7a3579cba1d · report fixture: signature base_str:str/0/str · returned str · level recorded 2026-09-25
guess_image_type_from_base64 evaluation/QvQ_Max.py ran fingerprinted Apache-2.0 (permissive) · d3ece86e3359d0bb · report fixture: signature base_str:str/0/str · returned str · level recorded 2026-09-25
load_dataset in 276 papers evaluation/o4-mini.py ran Apache-2.0 (permissive) · 59c478e3cad8958d · report fixture: signature file_path:str/0/str · returned dict · level recorded 2026-09-25
process_hades_data utils/hades.py ran Apache-2.0 (permissive) · 5a87ea6c5cf68cbc · report fixture: signature hades_root:str/0/str · returned list · level recorded 2026-09-25
read_mm_safebench utils/mmsafebench.py ran Apache-2.0 (permissive) · adb7a4c52bd43387 · report fixture: signature dataset_path:str/0/str, img_type:str/0/str, query_type:str/0/str · returned list · level recorded 2026-09-25
visualize_attention qwenmask.py ran Apache-2.0 (permissive) · 4f5467687896e741 · report fixture: signature image_input:Image/0/?, attention_tensor:ndarray/1/float, block_size:int/0/int, output_dir:str/0/str, color:str/0/str · returned dict · level recorded 2026-09-25
generate_response in 23 papers evaluation/gemini.py unverified Apache-2.0 (permissive) · 85fe3245f510864a · report
generate_response in 23 papers evaluation/gpt4o.py unverified Apache-2.0 (permissive) · a7841c5371a659a2 · report
load_model in 479 papers qwenmask.py unverified Apache-2.0 (permissive) · 317d970e6d4be3bc · report

Tasks

Visual Reasoning

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Methods

AttentionSoftmax

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections