ICO: Enhancing Semantic-Shift Jailbreaks via Iterative Context Optimization

arXiv:2608.03210 · cs.CL · Submitted 2026-08-04 · Read on arXiv

Hujian Zhu, Yihao Huang, Felix Juefei-Xu, Xinfeng Li, Peng Zeng, Simeng Qin, Qing Guo, Geguang Pu

cs.CL

Submitted: 2026-08-04

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Terminology

• Foundation Model — 14× in this paper · explained in 3 episodes
• Foundation Models — 14× in this paper · explained in 8 episodes
• Iterative Context Optimization — 9× in this paper
• Benchmark — 3× in this paper · explained in 2 episodes
• Multimodal Inputs — 2× in this paper · explained in 2 episodes
• Vision-language models — 2× in this paper · explained in 2 episodes
• Ablation — 1× in this paper · explained in 1 episode
• Ablation Study — 1× in this paper · explained in 3 episodes
• Adversarial Examples — 1× in this paper · explained in 2 episodes
• Composition — 1× in this paper · explained in 2 episodes
• Core — 1× in this paper · explained in 2 episodes
• Evaluation Framework — 1× in this paper · explained in 3 episodes
• Generalization — 1× in this paper · explained in 14 episodes
• Instruction Tuning — 1× in this paper · explained in 3 episodes
• Language Models (LLMs) — 1× in this paper · explained in 2 episodes
• Localization — 1× in this paper · explained in 2 episodes
• Multimodal Reasoning — 1× in this paper · explained in 5 episodes
• Prompt Engineering — 1× in this paper · explained in 6 episodes
• Semantic Gap — 1× in this paper · explained in 2 episodes

Sources

• On the Opportunities and Risks of Foundation Models
• Visual Instruction Tuning
• Red Teaming Language Models to Reduce Harms: Methods, Scaling Behaviors, and Lessons Learned
• In-Context Representation Hijacking
• Jailbreaking Vision-Language Models Through the Visual Modality
• GPT-4 Technical Report
• Gemini 1.5: Unlocking multimodal understanding across millions of tokens of context
• DeepSeek-V3 Technical Report
• The Llama 3 Herd of Models
• Qwen3-VL Technical Report
• Jailbreaking ChatGPT via Prompt Engineering: An Empirical Study
• Universal and Transferable Adversarial Attacks on Aligned Language Models
• Jailbreaking Black Box Large Language Models in Twenty Queries
• DeepSeek-V3.2: Pushing the Frontier of Open Large Language Models
• MM-SafetyBench: A Benchmark for Safety Evaluation of Multimodal Large Language Models

Related papers

• Exploring Solution Divergence and Its Effect on Large Language Model Problem Solving
• Ishigaki-IDS-Bench: A Benchmark for Generating Information Delivery Specification from BIM Information Requirements
• Subliminal Steering: Stronger Encoding of Hidden Signals
• MedStruct-S: A Benchmark for Key Discovery, Key-Conditioned QA and Semi-Structured Extraction from OCR Clinical Reports
• The End of Transformers? On Challenging Attention and the Rise of Sub-Quadratic Architectures
• Untangling the Mechanisms of Misleading Context in Medical Question Answering