Paper Citation Record · LEDGER

Universal Adversarial Attack on Aligned Multimodal LLMs

As of 18 August 2026, this Paper Citation Record lists 45 of 45 outbound references and 2 inbound Pith citation observations for arXiv:2502.07987.

A citation records a reference. It does not transfer a finding from one paper to another.

pith.paper-citation-record.v1
2502.07987 v3

Coverage vector

measured 45 of 45 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-08-08T11:17:01.831710Z

measured 47 of 47 standing notices

One-hop event checks from named stored sources.

Source: scholarly_work_events, retraction_status_cache, observed 2026-08-17T06:30:58.91139+00:00

measured 2 of 2 inbound itemization

Pith citing papers itemized under the disclosed page cap.

Source: paper_references, paper_reference_links, observed 2026-05-17T04:15:38.632039Z

measured 0 of 1 external citation measurements

A source-named dated measurement, never combined with another source.

Source: arxiv_reference, observed 2026-05-17T04:19:00.680668Z

Reference resolution

45 of 45 outbound references displayed

verified exact4
verified fuzzy0
unresolved41
parse uncertain0
malformed identifier0
metadata mismatch0

External citation measurements

No source-named external measurement is stored.

Outbound references

Observation 83947996-04f4-4e25-a36d-baf4911657c2 · outbound

This paper cites online" 'onlinestring :=.

Universal Adversarial Attack on Aligned Multimodal LLMs online" 'onlinestring :=

Reference 1

Resolution: unresolved; no resolver link, observed 2026-08-08T11:17:01.526225Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 6b6d5c79-8967-4e8e-9002-2915490634ed · outbound

This paper cites write newline.

Universal Adversarial Attack on Aligned Multimodal LLMs write newline

Reference 2

Resolution: unresolved; no resolver link, observed 2026-08-08T11:17:01.531793Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 996f2071-832e-41f8-a5ff-3288471d2623 · outbound

This paper cites Phi-3 Technical Report: A Highly Capable Language Model Locally on Your Phone.

Universal Adversarial Attack on Aligned Multimodal LLMs Phi-3 Technical Report: A Highly Capable Language Model Locally on Your Phone

Reference 3

Resolution: unresolved; no resolver link, observed 2026-08-08T11:17:01.536814Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 109c4f58-ee20-42cb-97d9-ed2e6bff3762 · outbound

This paper cites an unresolved cited work.

Universal Adversarial Attack on Aligned Multimodal LLMs Unresolved cited work

Reference 4

Resolution: unresolved; no resolver link, observed 2026-08-08T11:17:01.541668Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 109dba70-18a6-4f8d-812e-e1ebcfa6663c · outbound

This paper cites an unresolved cited work.

Universal Adversarial Attack on Aligned Multimodal LLMs Unresolved cited work

Reference 5

Resolution: unresolved; no resolver link, observed 2026-08-08T11:17:01.545940Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 87d96075-48ce-417a-a5d6-07c92ebf8825 · outbound

This paper cites an unresolved cited work.

Universal Adversarial Attack on Aligned Multimodal LLMs Unresolved cited work

Reference 6

Resolution: unresolved; raw_fallback, observed 2026-08-08T11:17:02.469027Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 2190d16d-822a-4cb5-b0d2-23090ffdb6e2 · outbound

This paper cites Hearing Shapes via p-Adic Laplacians.

Universal Adversarial Attack on Aligned Multimodal LLMs Hearing Shapes via p-Adic Laplacians

Reference 7

Resolution: verified exact; local_arxiv, observed 2026-08-08T11:17:02.344403Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 17d1a7cf-e579-4cbb-b1dd-91f41bbf9764 · outbound

This paper cites Llama Guard 3 Vision: Safeguarding Human-AI Image Understanding Conversations.

Universal Adversarial Attack on Aligned Multimodal LLMs Llama Guard 3 Vision: Safeguarding Human-AI Image Understanding Conversations

Reference 8

Resolution: unresolved; no resolver link, observed 2026-08-08T11:17:01.558974Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 32567272-1931-4005-a1b7-e44f6576d82e · outbound

This paper cites MM-REACT: Prompting ChatGPT for Multimodal Reasoning and Action.

Universal Adversarial Attack on Aligned Multimodal LLMs MM-REACT: Prompting ChatGPT for Multimodal Reasoning and Action

Reference 9

Resolution: unresolved; no resolver link, observed 2026-08-08T11:17:01.564096Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 23ad244c-07b0-4d5f-9f76-0268cb8e4009 · outbound

This paper cites Boosting Adversarial Attacks with Momentum.

Universal Adversarial Attack on Aligned Multimodal LLMs Boosting Adversarial Attacks with Momentum

Reference 10

Resolution: unresolved; no resolver link, observed 2026-08-08T11:17:01.568844Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 5070495d-a696-4e3b-a55c-2f5e606318f4 · outbound

This paper cites The Llama 3 Herd of Models.

Universal Adversarial Attack on Aligned Multimodal LLMs The Llama 3 Herd of Models

Reference 11

Resolution: unresolved; no resolver link, observed 2026-08-08T11:17:01.573464Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 8a28ffb6-f22f-4ade-aba3-4914e6ba9bfe · outbound

This paper cites FigStep: Jailbreaking Large Vision-Language Models via Typographic Visual Prompts.

Universal Adversarial Attack on Aligned Multimodal LLMs FigStep: Jailbreaking Large Vision-Language Models via Typographic Visual Prompts

Reference 12

Resolution: unresolved; no resolver link, observed 2026-08-08T11:17:01.577794Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e168420d-133e-496c-addd-12cea8c905d2 · outbound

This paper cites Agent Smith: A Single Image Can Jailbreak One Million Multimodal LLM Agents Exponentially Fast.

Universal Adversarial Attack on Aligned Multimodal LLMs Agent Smith: A Single Image Can Jailbreak One Million Multimodal LLM Agents Exponentially Fast

Reference 13

Resolution: unresolved; no resolver link, observed 2026-08-08T11:17:01.583853Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 4e6df700-2e4b-420b-9ba3-1f0f723e8b69 · outbound

This paper cites Gradient-based Adversarial Attacks against Text Transformers.

Universal Adversarial Attack on Aligned Multimodal LLMs Gradient-based Adversarial Attacks against Text Transformers

Reference 14

Resolution: unresolved; no resolver link, observed 2026-08-08T11:17:01.588766Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d7b17c01-f97d-47a8-9e84-3273d8c500a9 · outbound

This paper cites On the Trustworthiness of Generative Foundation Models: Guideline, Assessment, and Perspective.

Universal Adversarial Attack on Aligned Multimodal LLMs On the Trustworthiness of Generative Foundation Models: Guideline, Assessment, and Perspective

Reference 15

Resolution: unresolved; no resolver link, observed 2026-08-08T11:17:01.593548Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 3a7b6563-8297-4968-ba73-803a0947a0f4 · outbound

This paper cites an unresolved cited work.

Universal Adversarial Attack on Aligned Multimodal LLMs Unresolved cited work

Reference 16

Resolution: unresolved; no resolver link, observed 2026-08-08T11:17:01.598626Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 29bc14b6-a180-4c1f-be2a-f8f134fc2e9a · outbound

This paper cites Adversarial examples in the physical world.

Universal Adversarial Attack on Aligned Multimodal LLMs Adversarial examples in the physical world

Reference 17

Resolution: unresolved; no resolver link, observed 2026-08-08T11:17:01.603196Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 02e6487d-713d-474f-a74e-70382d7182c1 · outbound

This paper cites MapperGPT: Large Language Models for Linking and Mapping Entities.

Universal Adversarial Attack on Aligned Multimodal LLMs MapperGPT: Large Language Models for Linking and Mapping Entities

Reference 18

Resolution: verified exact; local_arxiv, observed 2026-08-08T11:17:02.205156Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 7b390dee-a4e2-4152-98a3-03b84154d200 · outbound

This paper cites Multiuser Resource Allocation for Semantic-Relay-Aided Text Transmissions.

Universal Adversarial Attack on Aligned Multimodal LLMs Multiuser Resource Allocation for Semantic-Relay-Aided Text Transmissions

Reference 19

Resolution: unresolved; no resolver link, observed 2026-08-08T11:17:01.613090Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 98a38884-cb80-4a7c-88ad-8b8316838b43 · outbound

This paper cites TextBugger: Generating Adversarial Text Against Real-world Applications.

Universal Adversarial Attack on Aligned Multimodal LLMs TextBugger: Generating Adversarial Text Against Real-world Applications

Reference 20

Resolution: unresolved; no resolver link, observed 2026-08-08T11:17:01.618425Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d5c48b04-37e5-45a9-bdb2-12075fdccf43 · outbound

This paper cites an unresolved cited work.

Universal Adversarial Attack on Aligned Multimodal LLMs Unresolved cited work

Reference 21

Resolution: unresolved; raw_fallback, observed 2026-08-08T11:17:02.447021Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 944b41db-afbb-4b61-a489-1b1cbbdd1221 · outbound

This paper cites an unresolved cited work.

Universal Adversarial Attack on Aligned Multimodal LLMs Unresolved cited work

Reference 22

Resolution: verified exact; doi, observed 2026-08-08T11:17:01.939365Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation ec4694b8-8ef6-4ce1-b7aa-632e13e9f2db · outbound

This paper cites an unresolved cited work.

Universal Adversarial Attack on Aligned Multimodal LLMs Unresolved cited work

Reference 23

Resolution: unresolved; no resolver link, observed 2026-08-08T11:17:01.635636Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 66b2a7e3-3117-4657-9023-6d592449408f · outbound

This paper cites an unresolved cited work.

Universal Adversarial Attack on Aligned Multimodal LLMs Unresolved cited work

Reference 24

Resolution: unresolved; no resolver link, observed 2026-08-08T11:17:01.642768Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 952d636f-32e5-4444-bf2e-33fb637c2390 · outbound

This paper cites an unresolved cited work.

Universal Adversarial Attack on Aligned Multimodal LLMs Unresolved cited work

Reference 25

Resolution: unresolved; no resolver link, observed 2026-08-08T11:17:01.650328Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f7fb9fc0-da95-4b8a-b745-96f5d8022a25 · outbound

This paper cites Visual-RolePlay: Universal Jailbreak Attack on MultiModal Large Language Models via Role-playing Image Character.

Universal Adversarial Attack on Aligned Multimodal LLMs Visual-RolePlay: Universal Jailbreak Attack on MultiModal Large Language Models via Role-playing Image Character

Reference 26

Resolution: unresolved; no resolver link, observed 2026-08-08T11:17:01.654636Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 0c736bd0-5358-47bf-bfb5-ce75fcc33230 · outbound

This paper cites Universal adversarial perturbations.

Universal Adversarial Attack on Aligned Multimodal LLMs Universal adversarial perturbations

Reference 27

Resolution: unresolved; no resolver link, observed 2026-08-08T11:17:01.659359Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 91100ab9-a510-47be-b52d-59235c18ba6b · outbound

This paper cites Adversarial Reprogramming of Text Classification Neural Networks.

Universal Adversarial Attack on Aligned Multimodal LLMs Adversarial Reprogramming of Text Classification Neural Networks

Reference 28

Resolution: unresolved; no resolver link, observed 2026-08-08T11:17:01.663868Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 4740269b-d406-4648-b9b6-1501d37f8428 · outbound

This paper cites Transferability in Machine Learning: from Phenomena to Black-Box Attacks using Adversarial Samples.

Universal Adversarial Attack on Aligned Multimodal LLMs Transferability in Machine Learning: from Phenomena to Black-Box Attacks using Adversarial Samples

Reference 29

Resolution: unresolved; no resolver link, observed 2026-08-08T11:17:01.668357Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 08fdb8bf-d39d-44ee-8847-db4b383435d5 · outbound

This paper cites Visual Adversarial Examples Jailbreak Aligned Large Language Models.

Universal Adversarial Attack on Aligned Multimodal LLMs Visual Adversarial Examples Jailbreak Aligned Large Language Models

Reference 30

Resolution: unresolved; no resolver link, observed 2026-08-08T11:17:01.672767Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 8947be28-0ed6-4562-b3fa-b12be27cd4c7 · outbound

This paper cites an unresolved cited work.

Universal Adversarial Attack on Aligned Multimodal LLMs Unresolved cited work

Reference 31

Resolution: unresolved; no resolver link, observed 2026-08-08T11:17:01.677370Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 780e3805-7313-4a42-8263-d1977ff01bf1 · outbound

This paper cites Jailbreak in pieces: Compositional Adversarial Attacks on Multi-Modal Language Models.

Universal Adversarial Attack on Aligned Multimodal LLMs Jailbreak in pieces: Compositional Adversarial Attacks on Multi-Modal Language Models

Reference 32

Resolution: unresolved; no resolver link, observed 2026-08-08T11:17:01.681810Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 8b83f0c6-28b2-43b2-88ce-baf6873700af · outbound

This paper cites Intriguing properties of neural networks.

Universal Adversarial Attack on Aligned Multimodal LLMs Intriguing properties of neural networks

Reference 33

Resolution: unresolved; no resolver link, observed 2026-08-08T11:17:01.686151Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 6c42bb88-5bcf-4288-9a99-d8cb2fec011b · outbound

This paper cites Gemma 3 Technical Report.

Universal Adversarial Attack on Aligned Multimodal LLMs Gemma 3 Technical Report

Reference 34

Resolution: unresolved; no resolver link, observed 2026-08-08T11:17:01.690284Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 0a81a350-d305-4711-ad03-6cfc9b6a80b3 · outbound

This paper cites an unresolved cited work.

Universal Adversarial Attack on Aligned Multimodal LLMs Unresolved cited work

Reference 35

Resolution: unresolved; no resolver link, observed 2026-08-08T11:17:01.695069Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d5deb967-4af0-4aae-ba2f-2045e62dfd03 · outbound

This paper cites Universal Adversarial Triggers for Attacking and Analyzing NLP.

Universal Adversarial Attack on Aligned Multimodal LLMs Universal Adversarial Triggers for Attacking and Analyzing NLP

Reference 36

Resolution: unresolved; no resolver link, observed 2026-08-08T11:17:01.699125Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation daea1bbc-d260-462a-a271-7d8ac90e2d65 · outbound

This paper cites Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution.

Universal Adversarial Attack on Aligned Multimodal LLMs Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution

Reference 37

Resolution: unresolved; no resolver link, observed 2026-08-08T11:17:01.703623Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 0b6b6ce2-e9a8-46c6-80b3-923d950082c2 · outbound

This paper cites Jailbroken: How Does LLM Safety Training Fail?.

Universal Adversarial Attack on Aligned Multimodal LLMs Jailbroken: How Does LLM Safety Training Fail?

Reference 38

Resolution: unresolved; no resolver link, observed 2026-08-08T11:17:01.716035Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 1f32d48a-0c5b-4d44-af44-097b80cd8401 · outbound

This paper cites Can audio-visual integration strengthen robustness under multimodal attacks?.

Universal Adversarial Attack on Aligned Multimodal LLMs Can audio-visual integration strengthen robustness under multimodal attacks?

Reference 39

Resolution: verified exact; local_arxiv, observed 2026-08-08T11:17:02.002221Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 9f69df66-1b14-40c5-9abc-d4c005090d4f · outbound

This paper cites an unresolved cited work.

Universal Adversarial Attack on Aligned Multimodal LLMs Unresolved cited work

Reference 40

Resolution: unresolved; raw_fallback, observed 2026-08-08T11:17:02.398263Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 70b85627-294c-4a45-b4a4-3a8fbafb761f · outbound

This paper cites SafeBench: A Safety Evaluation Framework for Multimodal Large Language Models.

Universal Adversarial Attack on Aligned Multimodal LLMs SafeBench: A Safety Evaluation Framework for Multimodal Large Language Models

Reference 41

Resolution: unresolved; no resolver link, observed 2026-08-08T11:17:01.814133Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e9958c13-871e-4078-9ef3-8abb2e49dfcf · outbound

This paper cites an unresolved cited work.

Universal Adversarial Attack on Aligned Multimodal LLMs Unresolved cited work

Reference 42

Resolution: unresolved; raw_fallback, observed 2026-08-08T11:17:02.384817Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 42fe3565-6180-41c2-89c9-b0608923dfc9 · outbound

This paper cites an unresolved cited work.

Universal Adversarial Attack on Aligned Multimodal LLMs Unresolved cited work

Reference 43

Resolution: unresolved; raw_fallback, observed 2026-08-08T11:17:02.370972Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 27bc654c-76ad-449c-90da-aa545e47a784 · outbound

This paper cites Parameter-Efficient Orthogonal Finetuning via Butterfly Factorization.

Universal Adversarial Attack on Aligned Multimodal LLMs Parameter-Efficient Orthogonal Finetuning via Butterfly Factorization

Reference 44

Resolution: unresolved; no resolver link, observed 2026-08-08T11:17:01.827247Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 8fb4450b-886d-4d54-b91f-33826dc4e87e · outbound

This paper cites Universal and Transferable Adversarial Attacks on Aligned Language Models.

Universal Adversarial Attack on Aligned Multimodal LLMs Universal and Transferable Adversarial Attacks on Aligned Language Models

Reference 45

Resolution: unresolved; no resolver link, observed 2026-08-08T11:17:01.831710Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Pith citing papers

Observation 6790d80d-ccf0-4bfb-88a8-452029705eb7 · inbound

Breaking the Illusion: Consensus-Based Generative Mitigation of Adversarial Illusions in Multi-Modal Embeddings cites this paper.

Breaking the Illusion: Consensus-Based Generative Mitigation of Adversarial Illusions in Multi-Modal Embeddings Universal Adversarial Attack on Aligned Multimodal LLMs

Reference 21

Resolution: verified exact; arxiv_id, observed 2026-05-17T04:19:00.683342Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.

Observation 49a54231-3544-4055-9c6b-11b542bcc400 · inbound

VisInject: Disruption != Injection -- A Dual-Dimension Evaluation of Universal Adversarial Attacks on Vision-Language Models cites this paper.

VisInject: Disruption != Injection -- A Dual-Dimension Evaluation of Universal Adversarial Attacks on Vision-Language Models Universal Adversarial Attack on Aligned Multimodal LLMs

Reference 25

Resolution: verified exact; arxiv_id, observed 2026-05-11T16:56:08.033580Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.