Jailbreak Foundry: From Papers to Runnable Attacks for Reproducible Benchmarking
cs.CR, cs.AI, cs.CL, cs.LG
Submitted: 2026-02-27
Updated: 2026-07-17
Journal ref: Proceedings of the 43rd International Conference on Machine Learning, PMLR 306:29548-29590, 2026
Code: https://github.com/OpenSQZ/Jailbreak-Foundry
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Terminology
Sources
- Between a Rock and a Hard Place: The Tension Between Ethical Reasoning and Safety Alignment in LLMs
- Friend or Foe: How LLMs' Safety Mind Gets Fooled by Intent Shift Attack
- The Llama 3 Herd of Models
- GuidedBench: Measuring and Mitigating the Evaluation Discrepancies of In-the-wild LLM Jailbreak Methods
- Alphabet Index Mapping: Jailbreaking LLMs through Semantic Dissimilarity
- DeepInception: Hypnotize Large Language Model to Be Jailbreaker
- EquaCode: A Multi-Strategy Jailbreak Approach for Large Language Models via Equation Solving and Code Completion
- Reasoning as a Weapon: Adaptive Dual-Path Jailbreak Attack on Large Language Models
- Response Attack: Exploiting Contextual Priming to Jailbreak Large Language Models
- gpt-oss-120b & gpt-oss-20b Model Card
- GPT-4 Technical Report
- OpenRT: An Open-Source Red Teaming Framework for Multimodal LLMs
- You Know What I'm Saying: Jailbreak Attack via Implicit Reference
- SciReplicate-Bench: Benchmarking LLMs in Agent-driven Algorithmic Reproduction from Research Papers
- Qwen3 Technical Report
- EasyJailbreak: A Unified Framework for Jailbreaking Large Language Models
- Universal and Transferable Adversarial Attacks on Aligned Language Models
Related papers
- SoK: AI-Augmented Binary Reversing
- Relaxed Sender Anonymity for CBDC Interbank Settlement: A Zero-Knowledge Approach on Permissioned EVM
- Calibration-Family Overfit: Why Trusted Sabotage Monitors Don't Transfer Across Lineages
- Efficient Fuzzy PSI under One-Sided Assumptions
- Sealing the Audit-Runtime Gap for LLM Skills
- Token Composition: A Graph Based on EVM Logs