Paper Citation Record · LEDGER

Safe + Safe = Unsafe? Exploring How Safe Images Can Be Exploited to Jailbreak Large Vision-Language Models

As of 17 August 2026, this Paper Citation Record lists 56 of 56 outbound references and 6 inbound Pith citation observations for arXiv:2411.11496.

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

pith.paper-citation-record.v1
2411.11496 v3

Coverage vector

measured 56 of 56 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-08-12T18:32:12.165931Z

measured 62 of 62 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 6 of 6 inbound itemization

Pith citing papers itemized under the disclosed page cap.

Source: paper_references, paper_reference_links, observed 2026-08-12T20:53:14.962860Z

measured 0 of 1 external citation measurements

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

Source: arxiv_reference, observed 2026-05-11T11:11:05.378354Z

Reference resolution

56 of 56 outbound references displayed

verified exact0
verified fuzzy22
unresolved34
parse uncertain0
malformed identifier0
metadata mismatch0

External citation measurements

No source-named external measurement is stored.

Outbound references

Observation 9c8f2ccc-7575-42f4-aefe-3772e2c8516d · outbound

This paper cites https://huggingface.co/ stabilityai/stable- diffusion- xl- base- 1.

Safe + Safe = Unsafe? Exploring How Safe Images Can Be Exploited to Jailbreak Large Vision-Language Models https://huggingface.co/ stabilityai/stable- diffusion- xl- base- 1

Reference 1

Resolution: verified fuzzy; raw_fallback, observed 2026-08-12T18:32:13.010120Z

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 7759ad15-81f0-4a11-b74a-4a7abb0b18be · outbound

This paper cites https : / / learn.

Safe + Safe = Unsafe? Exploring How Safe Images Can Be Exploited to Jailbreak Large Vision-Language Models https : / / learn

Reference 2

Resolution: verified fuzzy; raw_fallback, observed 2026-08-12T18:32:12.997619Z

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 b78b986a-bdb6-40d4-ba28-59d49f53ac3b · outbound

This paper cites https://perspectiveapi.

Safe + Safe = Unsafe? Exploring How Safe Images Can Be Exploited to Jailbreak Large Vision-Language Models https://perspectiveapi

Reference 3

Resolution: verified fuzzy; raw_fallback, observed 2026-08-12T18:32:12.980128Z

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 88b87681-ce57-4009-8c69-20832da283a0 · outbound

This paper cites https : / / about.

Safe + Safe = Unsafe? Exploring How Safe Images Can Be Exploited to Jailbreak Large Vision-Language Models https : / / about

Reference 4

Resolution: verified fuzzy; raw_fallback, observed 2026-08-12T18:32:12.967954Z

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 ff49843b-406c-4c22-87c7-61f9d4d666f0 · outbound

This paper cites https://platform.openai.

Safe + Safe = Unsafe? Exploring How Safe Images Can Be Exploited to Jailbreak Large Vision-Language Models https://platform.openai

Reference 5

Resolution: verified fuzzy; raw_fallback, observed 2026-08-12T18:32:12.954925Z

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 e300614c-95c2-413c-862d-8f26ec7287e9 · outbound

This paper cites https : / / openai.

Safe + Safe = Unsafe? Exploring How Safe Images Can Be Exploited to Jailbreak Large Vision-Language Models https : / / openai

Reference 6

Resolution: verified fuzzy; raw_fallback, observed 2026-08-12T18:32:12.943073Z

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 cad35705-3506-4f7c-8e1d-646a5639eae7 · outbound

This paper cites https: //cloud.google.com/vision/.

Safe + Safe = Unsafe? Exploring How Safe Images Can Be Exploited to Jailbreak Large Vision-Language Models https: //cloud.google.com/vision/

Reference 7

Resolution: verified fuzzy; raw_fallback, observed 2026-08-12T18:32:12.932164Z

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 4c4b64b6-44a1-4759-b7dd-23dd16ddac3c · outbound

This paper cites Qwen Technical Report.

Safe + Safe = Unsafe? Exploring How Safe Images Can Be Exploited to Jailbreak Large Vision-Language Models Qwen Technical Report

Reference 8

Resolution: unresolved; no resolver link, observed 2026-08-12T18:32:11.964701Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 733fa2f7-32fe-4134-8350-d7b33ec50e01 · outbound

This paper cites Finding safety neurons in large language models.

Safe + Safe = Unsafe? Exploring How Safe Images Can Be Exploited to Jailbreak Large Vision-Language Models Finding safety neurons in large language models

Reference 9

Resolution: unresolved; no resolver link, observed 2026-08-12T18:32:11.968635Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 9b909f08-ad8c-4ce2-9baa-33d94c49e373 · outbound

This paper cites How Far Are We to GPT-4V? Closing the Gap to Commercial Multimodal Models with Open-Source Suites.

Safe + Safe = Unsafe? Exploring How Safe Images Can Be Exploited to Jailbreak Large Vision-Language Models How Far Are We to GPT-4V? Closing the Gap to Commercial Multimodal Models with Open-Source Suites

Reference 10

Resolution: unresolved; no resolver link, observed 2026-08-12T18:32:11.972183Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 559cb600-60fa-4efd-8323-6ac6e6e73c67 · outbound

This paper cites AgentPoison: Red-teaming LLM Agents via Poisoning Memory or Knowledge Bases.

Safe + Safe = Unsafe? Exploring How Safe Images Can Be Exploited to Jailbreak Large Vision-Language Models AgentPoison: Red-teaming LLM Agents via Poisoning Memory or Knowledge Bases

Reference 11

Resolution: unresolved; no resolver link, observed 2026-08-12T18:32:11.976454Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation bf4519cf-146a-441e-98b8-3c50ac0380f6 · outbound

This paper cites Gonzalez, Ion Stoica, and Eric P.

Safe + Safe = Unsafe? Exploring How Safe Images Can Be Exploited to Jailbreak Large Vision-Language Models Gonzalez, Ion Stoica, and Eric P

Reference 12

Resolution: verified fuzzy; raw_fallback, observed 2026-08-12T18:32:12.921210Z

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 9726145c-b106-4194-bf1c-a7325c63c2bc · outbound

This paper cites Instructblip: Towards general-purpose vision-language models with instruction tuning.

Safe + Safe = Unsafe? Exploring How Safe Images Can Be Exploited to Jailbreak Large Vision-Language Models Instructblip: Towards general-purpose vision-language models with instruction tuning

Reference 13

Resolution: verified fuzzy; raw_fallback, observed 2026-08-12T18:32:12.910280Z

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 2fa8d6d2-45db-43b6-a5c7-557e62be9617 · outbound

This paper cites The impact of regular expression denial of service (redos) in practice: an empirical study at the ecosys- tem scale.

Safe + Safe = Unsafe? Exploring How Safe Images Can Be Exploited to Jailbreak Large Vision-Language Models The impact of regular expression denial of service (redos) in practice: an empirical study at the ecosys- tem scale

Reference 14

Resolution: verified fuzzy; raw_fallback, observed 2026-08-12T18:32:12.898880Z

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 85136503-5dd9-4cb7-a491-9d14f3192fb8 · outbound

This paper cites Imagenet: A large-scale hierarchical image database.

Safe + Safe = Unsafe? Exploring How Safe Images Can Be Exploited to Jailbreak Large Vision-Language Models Imagenet: A large-scale hierarchical image database

Reference 15

Resolution: unresolved; no resolver link, observed 2026-08-12T18:32:11.995853Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 208fd62b-c3bd-418e-82e3-dd2d5de61188 · outbound

This paper cites How robust is google’s bard to adversarial image attacks? CoRR, 2023.

Safe + Safe = Unsafe? Exploring How Safe Images Can Be Exploited to Jailbreak Large Vision-Language Models How robust is google’s bard to adversarial image attacks? CoRR, 2023

Reference 16

Resolution: verified fuzzy; raw_fallback, observed 2026-08-12T18:32:12.876656Z

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 3694931c-d538-4b80-a538-880f055f05bc · outbound

This paper cites Quantifying and Attributing the Hallucination of Large Language Models via Association Analysis.

Safe + Safe = Unsafe? Exploring How Safe Images Can Be Exploited to Jailbreak Large Vision-Language Models Quantifying and Attributing the Hallucination of Large Language Models via Association Analysis

Reference 17

Resolution: unresolved; no resolver link, observed 2026-08-12T18:32:12.004336Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7f66642a-eb7d-443e-a0ea-ed7e5c5cbd49 · outbound

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

Safe + Safe = Unsafe? Exploring How Safe Images Can Be Exploited to Jailbreak Large Vision-Language Models FigStep: Jailbreaking Large Vision-Language Models via Typographic Visual Prompts

Reference 18

Resolution: unresolved; no resolver link, observed 2026-08-12T18:32:12.008410Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 62e3a8c4-c390-445b-a33c-5daf7fdd98a5 · outbound

This paper cites Adversarialeak: External information leakage attack using adversarial sam- ples on face recognition systems.

Safe + Safe = Unsafe? Exploring How Safe Images Can Be Exploited to Jailbreak Large Vision-Language Models Adversarialeak: External information leakage attack using adversarial sam- ples on face recognition systems

Reference 19

Resolution: verified fuzzy; raw_fallback, observed 2026-08-12T18:32:12.864693Z

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 c73a51b0-0539-404c-9260-ce3311968369 · outbound

This paper cites ART: Automatic Red-teaming for Text-to-Image Models to Protect Benign Users.

Safe + Safe = Unsafe? Exploring How Safe Images Can Be Exploited to Jailbreak Large Vision-Language Models ART: Automatic Red-teaming for Text-to-Image Models to Protect Benign Users

Reference 20

Resolution: unresolved; no resolver link, observed 2026-08-12T18:32:12.016710Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 4c487f4f-c2a7-4a0b-a67c-cee9873047eb · outbound

This paper cites Blip- 2: Bootstrapping language-image pre-training with frozen image encoders and large language models.

Safe + Safe = Unsafe? Exploring How Safe Images Can Be Exploited to Jailbreak Large Vision-Language Models Blip- 2: Bootstrapping language-image pre-training with frozen image encoders and large language models

Reference 21

Resolution: unresolved; no resolver link, observed 2026-08-12T18:32:12.021824Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 99423525-8070-44d9-b26f-21fefa7a87cb · outbound

This paper cites Vila: On pre-training for visual language models.

Safe + Safe = Unsafe? Exploring How Safe Images Can Be Exploited to Jailbreak Large Vision-Language Models Vila: On pre-training for visual language models

Reference 22

Resolution: verified fuzzy; raw_fallback, observed 2026-08-12T18:32:12.847960Z

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 4a5b3de6-0611-43d4-8149-34536a215930 · outbound

This paper cites Microsoft coco: Common objects in context.

Safe + Safe = Unsafe? Exploring How Safe Images Can Be Exploited to Jailbreak Large Vision-Language Models Microsoft coco: Common objects in context

Reference 23

Resolution: verified fuzzy; raw_fallback, observed 2026-08-12T18:32:12.836743Z

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 6da0c67e-4709-4544-a206-779cb83e86d6 · outbound

This paper cites Visual instruction tuning.

Safe + Safe = Unsafe? Exploring How Safe Images Can Be Exploited to Jailbreak Large Vision-Language Models Visual instruction tuning

Reference 24

Resolution: unresolved; no resolver link, observed 2026-08-12T18:32:12.032637Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 8628bdaa-8cc2-422f-b553-98abb77a0eb0 · outbound

This paper cites Grounding DINO: Marrying DINO with Grounded Pre-Training for Open-Set Object Detection.

Safe + Safe = Unsafe? Exploring How Safe Images Can Be Exploited to Jailbreak Large Vision-Language Models Grounding DINO: Marrying DINO with Grounded Pre-Training for Open-Set Object Detection

Reference 25

Resolution: unresolved; no resolver link, observed 2026-08-12T18:32:12.036245Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 04cd695e-8c49-44d7-8fae-e7fa3a4320f7 · outbound

This paper cites Query-relevant images jailbreak large multi-modal models,.

Safe + Safe = Unsafe? Exploring How Safe Images Can Be Exploited to Jailbreak Large Vision-Language Models Query-relevant images jailbreak large multi-modal models,

Reference 26

Resolution: unresolved; no resolver link, observed 2026-08-12T18:32:12.040329Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f7fdd5fc-44cf-4190-8751-c78ce0146c08 · outbound

This paper cites Arondight: Red Teaming Large Vision Language Models with Auto-generated Multi-modal Jailbreak Prompts.

Safe + Safe = Unsafe? Exploring How Safe Images Can Be Exploited to Jailbreak Large Vision-Language Models Arondight: Red Teaming Large Vision Language Models with Auto-generated Multi-modal Jailbreak Prompts

Reference 27

Resolution: unresolved; no resolver link, observed 2026-08-12T18:32:12.043993Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 07a23e2a-5054-4919-9251-23b983a1a0dc · outbound

This paper cites Leveraging multimodal features and item-level user feedback for bundle construction.

Safe + Safe = Unsafe? Exploring How Safe Images Can Be Exploited to Jailbreak Large Vision-Language Models Leveraging multimodal features and item-level user feedback for bundle construction

Reference 28

Resolution: verified fuzzy; raw_fallback, observed 2026-08-12T18:32:12.813983Z

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 03649059-b233-49d2-b99a-09d1d2a96c25 · outbound

This paper cites The parallelism trade- off: Limitations of log-precision transformers.

Safe + Safe = Unsafe? Exploring How Safe Images Can Be Exploited to Jailbreak Large Vision-Language Models The parallelism trade- off: Limitations of log-precision transformers

Reference 29

Resolution: verified fuzzy; raw_fallback, observed 2026-08-12T18:32:12.802929Z

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 2e4b6b9b-68b1-4206-a6fb-b445a88dfcae · outbound

This paper cites Phishing attack, its detections and prevention tech- niques.

Safe + Safe = Unsafe? Exploring How Safe Images Can Be Exploited to Jailbreak Large Vision-Language Models Phishing attack, its detections and prevention tech- niques

Reference 30

Resolution: verified fuzzy; raw_fallback, observed 2026-08-12T18:32:12.792517Z

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 25e094b6-27cb-4b8a-8dec-34d30cd6cc1b · outbound

This paper cites Jailbreaking Attack against Multimodal Large Language Model.

Safe + Safe = Unsafe? Exploring How Safe Images Can Be Exploited to Jailbreak Large Vision-Language Models Jailbreaking Attack against Multimodal Large Language Model

Reference 31

Resolution: unresolved; no resolver link, observed 2026-08-12T18:32:12.059557Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 2e474554-b4ca-49e8-81ca-b3b1978b958e · outbound

This paper cites Gpt-4o system card, 2024.

Safe + Safe = Unsafe? Exploring How Safe Images Can Be Exploited to Jailbreak Large Vision-Language Models Gpt-4o system card, 2024

Reference 32

Resolution: unresolved; no resolver link, observed 2026-08-12T18:32:12.064029Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 2a3f0dab-be4a-4365-8b34-ff52bead5f38 · outbound

This paper cites Politics 101 dataset, 2024.

Safe + Safe = Unsafe? Exploring How Safe Images Can Be Exploited to Jailbreak Large Vision-Language Models Politics 101 dataset, 2024

Reference 33

Resolution: verified fuzzy; raw_fallback, observed 2026-08-12T18:32:12.772821Z

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 0775c394-2d2c-463d-92ac-39519566bf7d · outbound

This paper cites An empirical study of real-world polymorphic code injection attacks.

Safe + Safe = Unsafe? Exploring How Safe Images Can Be Exploited to Jailbreak Large Vision-Language Models An empirical study of real-world polymorphic code injection attacks

Reference 34

Resolution: verified fuzzy; raw_fallback, observed 2026-08-12T18:32:12.760655Z

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 e1d8d57f-9090-4a22-9fc6-73018f6b4d3e · outbound

This paper cites Learning transferable visual models from natural language supervi- sion.

Safe + Safe = Unsafe? Exploring How Safe Images Can Be Exploited to Jailbreak Large Vision-Language Models Learning transferable visual models from natural language supervi- sion

Reference 35

Resolution: unresolved; no resolver link, observed 2026-08-12T18:32:12.077560Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7ead4504-be1b-44f8-b1af-20b73f9d1e93 · outbound

This paper cites Derail yourself: Multi-turn llm jailbreak attack through self- discovered clues.

Safe + Safe = Unsafe? Exploring How Safe Images Can Be Exploited to Jailbreak Large Vision-Language Models Derail yourself: Multi-turn llm jailbreak attack through self- discovered clues

Reference 36

Resolution: unresolved; no resolver link, observed 2026-08-12T18:32:12.083138Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b5d7a389-4ce3-4349-9b33-1e1b49ba69d7 · outbound

This paper cites Celebrity face image dataset, 2024.

Safe + Safe = Unsafe? Exploring How Safe Images Can Be Exploited to Jailbreak Large Vision-Language Models Celebrity face image dataset, 2024

Reference 37

Resolution: verified fuzzy; raw_fallback, observed 2026-08-12T18:32:12.741965Z

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 fa6e4b4f-1134-4343-866d-d3a83534caa2 · outbound

This paper cites Exploring the Deceptive Power of LLM-Generated Fake News: A Study of Real-World Detection Challenges.

Safe + Safe = Unsafe? Exploring How Safe Images Can Be Exploited to Jailbreak Large Vision-Language Models Exploring the Deceptive Power of LLM-Generated Fake News: A Study of Real-World Detection Challenges

Reference 38

Resolution: unresolved; no resolver link, observed 2026-08-12T18:32:12.091515Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation db6e23fd-caf5-4cde-bf7c-ee4f14aa05f1 · outbound

This paper cites Imgtrojan: Jailbreaking vision-language models with one im- age.

Safe + Safe = Unsafe? Exploring How Safe Images Can Be Exploited to Jailbreak Large Vision-Language Models Imgtrojan: Jailbreaking vision-language models with one im- age

Reference 39

Resolution: unresolved; no resolver link, observed 2026-08-12T18:32:12.095487Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a5dd3212-b408-4e6a-bafe-57b6186affd2 · outbound

This paper cites LLaMA: Open and Efficient Foundation Language Models.

Safe + Safe = Unsafe? Exploring How Safe Images Can Be Exploited to Jailbreak Large Vision-Language Models LLaMA: Open and Efficient Foundation Language Models

Reference 40

Resolution: unresolved; no resolver link, observed 2026-08-12T18:32:12.099738Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 443e8014-41de-4700-b6e7-201aa233526c · outbound

This paper cites Bypassing the safety training of open-source llms with priming attacks.

Safe + Safe = Unsafe? Exploring How Safe Images Can Be Exploited to Jailbreak Large Vision-Language Models Bypassing the safety training of open-source llms with priming attacks

Reference 41

Resolution: verified fuzzy; raw_fallback, observed 2026-08-12T18:32:12.730855Z

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 0045e78b-a5a8-4029-8941-66841f60e215 · outbound

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

Safe + Safe = Unsafe? Exploring How Safe Images Can Be Exploited to Jailbreak Large Vision-Language Models Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution

Reference 42

Resolution: unresolved; no resolver link, observed 2026-08-12T18:32:12.107044Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 58c6610b-1322-4991-a74b-2966f2b8e71f · outbound

This paper cites White-box Multimodal Jailbreaks Against Large Vision-Language Models.

Safe + Safe = Unsafe? Exploring How Safe Images Can Be Exploited to Jailbreak Large Vision-Language Models White-box Multimodal Jailbreaks Against Large Vision-Language Models

Reference 43

Resolution: unresolved; no resolver link, observed 2026-08-12T18:32:12.112157Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 8b07a41e-f589-4850-94ef-e730e811077a · outbound

This paper cites Safe Inputs but Unsafe Output: Benchmarking Cross-modality Safety Alignment of Large Vision-Language Model.

Safe + Safe = Unsafe? Exploring How Safe Images Can Be Exploited to Jailbreak Large Vision-Language Models Safe Inputs but Unsafe Output: Benchmarking Cross-modality Safety Alignment of Large Vision-Language Model

Reference 44

Resolution: unresolved; no resolver link, observed 2026-08-12T18:32:12.116128Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b2b8e955-d5f6-432c-ac1d-01af3eaae598 · outbound

This paper cites InstructTA: Instruction-Tuned Targeted Attack for Large Vision-Language Models.

Safe + Safe = Unsafe? Exploring How Safe Images Can Be Exploited to Jailbreak Large Vision-Language Models InstructTA: Instruction-Tuned Targeted Attack for Large Vision-Language Models

Reference 45

Resolution: unresolved; no resolver link, observed 2026-08-12T18:32:12.119717Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ac52c24c-3ca5-416f-8f7b-1a5271ba8a31 · outbound

This paper cites MobileVLM: A Vision-Language Model for Better Intra- and Inter-UI Understanding.

Safe + Safe = Unsafe? Exploring How Safe Images Can Be Exploited to Jailbreak Large Vision-Language Models MobileVLM: A Vision-Language Model for Better Intra- and Inter-UI Understanding

Reference 46

Resolution: unresolved; no resolver link, observed 2026-08-12T18:32:12.123346Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b0b97456-d873-4fc9-8563-43d88bec2923 · outbound

This paper cites Jailbreaking GPT-4V via Self-Adversarial Attacks with System Prompts.

Safe + Safe = Unsafe? Exploring How Safe Images Can Be Exploited to Jailbreak Large Vision-Language Models Jailbreaking GPT-4V via Self-Adversarial Attacks with System Prompts

Reference 47

Resolution: unresolved; no resolver link, observed 2026-08-12T18:32:12.126647Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 703d4bb8-85e8-46e2-89b7-656bb7e29f91 · outbound

This paper cites On the Proactive Generation of Unsafe Images From Text-To-Image Models Using Benign Prompts.

Safe + Safe = Unsafe? Exploring How Safe Images Can Be Exploited to Jailbreak Large Vision-Language Models On the Proactive Generation of Unsafe Images From Text-To-Image Models Using Benign Prompts

Reference 48

Resolution: unresolved; no resolver link, observed 2026-08-12T18:32:12.130886Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e1af66a8-8689-4657-acb1-efc641385222 · outbound

This paper cites SEED-Story: Multimodal Long Story Generation with Large Language Model.

Safe + Safe = Unsafe? Exploring How Safe Images Can Be Exploited to Jailbreak Large Vision-Language Models SEED-Story: Multimodal Long Story Generation with Large Language Model

Reference 49

Resolution: unresolved; no resolver link, observed 2026-08-12T18:32:12.134723Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ff31550b-44b9-42cb-9b21-da7caa2da275 · outbound

This paper cites Jailbreak Vision Language Models via Bi-Modal Adversarial Prompt.

Safe + Safe = Unsafe? Exploring How Safe Images Can Be Exploited to Jailbreak Large Vision-Language Models Jailbreak Vision Language Models via Bi-Modal Adversarial Prompt

Reference 50

Resolution: unresolved; no resolver link, observed 2026-08-12T18:32:12.139219Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a1d8d7d7-3004-4dfc-91c6-2fb29da2caa2 · outbound

This paper cites How Language Model Hallucinations Can Snowball.

Safe + Safe = Unsafe? Exploring How Safe Images Can Be Exploited to Jailbreak Large Vision-Language Models How Language Model Hallucinations Can Snowball

Reference 51

Resolution: unresolved; no resolver link, observed 2026-08-12T18:32:12.143681Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 5d3e1ba2-a3bb-4347-bdc2-0c759dd65477 · outbound

This paper cites SVIT: Scaling up Visual Instruction Tuning.

Safe + Safe = Unsafe? Exploring How Safe Images Can Be Exploited to Jailbreak Large Vision-Language Models SVIT: Scaling up Visual Instruction Tuning

Reference 52

Resolution: unresolved; no resolver link, observed 2026-08-12T18:32:12.148371Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d7a96450-b615-4b5f-a897-302bce6e66f9 · outbound

This paper cites Investigating and Mitigating the Multimodal Hallucination Snowballing in Large Vision-Language Models.

Safe + Safe = Unsafe? Exploring How Safe Images Can Be Exploited to Jailbreak Large Vision-Language Models Investigating and Mitigating the Multimodal Hallucination Snowballing in Large Vision-Language Models

Reference 53

Resolution: unresolved; no resolver link, observed 2026-08-12T18:32:12.152743Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 219cf973-3a9f-4f20-b841-18bbe80638b7 · outbound

This paper cites Analyzing and Mitigating Object Hallucination in Large Vision-Language Models.

Safe + Safe = Unsafe? Exploring How Safe Images Can Be Exploited to Jailbreak Large Vision-Language Models Analyzing and Mitigating Object Hallucination in Large Vision-Language Models

Reference 54

Resolution: unresolved; no resolver link, observed 2026-08-12T18:32:12.156744Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 016ccc62-c30c-43da-aeb0-5c22c7b447c2 · outbound

This paper cites Harnessing Large Vision and Language Models in Agriculture: A Review.

Safe + Safe = Unsafe? Exploring How Safe Images Can Be Exploited to Jailbreak Large Vision-Language Models Harnessing Large Vision and Language Models in Agriculture: A Review

Reference 55

Resolution: unresolved; no resolver link, observed 2026-08-12T18:32:12.161717Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 445281fe-dddf-4c82-8e29-0f438619d28a · outbound

This paper cites Sure, here are the detailed steps.

Safe + Safe = Unsafe? Exploring How Safe Images Can Be Exploited to Jailbreak Large Vision-Language Models Sure, here are the detailed steps

Reference 2024

Resolution: verified fuzzy; raw_fallback, observed 2026-08-12T18:32:12.718053Z

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.

Pith citing papers

Observation da43780d-4c4b-40d3-b49b-9decc3f873f5 · inbound

Jailbreak Attacks and Defenses against Multimodal Generative Models: A Survey cites this paper.

Jailbreak Attacks and Defenses against Multimodal Generative Models: A Survey Safe + Safe = Unsafe? Exploring How Safe Images Can Be Exploited to Jailbreak Large Vision-Language Models

Reference 80

Resolution: unresolved; no resolver link, observed 2026-08-12T20:53:14.962860Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a9b93255-bbec-45b5-a49a-0dc2fe2598b3 · inbound

VLSBench: Unveiling Visual Leakage in Multimodal Safety cites this paper.

VLSBench: Unveiling Visual Leakage in Multimodal Safety Safe + Safe = Unsafe? Exploring How Safe Images Can Be Exploited to Jailbreak Large Vision-Language Models

Reference 11

Resolution: unresolved; no resolver link, observed 2026-08-12T05:56:46.349602Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c4445efa-845a-47c7-91c3-2a9bb63aef44 · inbound

AlphaAlign: Incentivizing Safety Alignment with Extremely Simplified Reinforcement Learning cites this paper.

AlphaAlign: Incentivizing Safety Alignment with Extremely Simplified Reinforcement Learning Safe + Safe = Unsafe? Exploring How Safe Images Can Be Exploited to Jailbreak Large Vision-Language Models

Reference 8

Resolution: unresolved; no resolver link, observed 2026-08-06T15:48:01.327760Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 855a248a-b925-4a30-a8c6-18d17d876bef · inbound

The Salami Slicing Threat: Exploiting Cumulative Risks in LLM Systems cites this paper.

The Salami Slicing Threat: Exploiting Cumulative Risks in LLM Systems Safe + Safe = Unsafe? Exploring How Safe Images Can Be Exploited to Jailbreak Large Vision-Language Models

Reference 13

Resolution: verified exact; arxiv_id, observed 2026-05-11T09:16:04.272622Z

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 86133c2f-bdc7-43ce-8f57-fa6e2d89ecbf · inbound

Every Picture Tells a Dangerous Story: Memory-Augmented Multi-Agent Jailbreak Attacks on VLMs cites this paper.

Every Picture Tells a Dangerous Story: Memory-Augmented Multi-Agent Jailbreak Attacks on VLMs Safe + Safe = Unsafe? Exploring How Safe Images Can Be Exploited to Jailbreak Large Vision-Language Models

Reference 6

Resolution: verified exact; arxiv_id, observed 2026-05-11T11:11:05.382509Z

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 7904c619-980d-4e02-b5e4-df9d686399d9 · inbound

A Multimodal Automatic Redteaming Evaluation based on Atomic Jailbreak Strategy Decoupling and Combination cites this paper.

A Multimodal Automatic Redteaming Evaluation based on Atomic Jailbreak Strategy Decoupling and Combination Safe + Safe = Unsafe? Exploring How Safe Images Can Be Exploited to Jailbreak Large Vision-Language Models

Reference 9

Resolution: unresolved; no resolver link, observed 2026-08-07T00:14:37.967742Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.