mlsanityData to 14 Oct 2026

Logic Jailbreak: Efficiently Unlocking LLM Safety Restrictions Through Formal Logical Expression

arXiv 2505.13527 ↗First posted 18 May 2025cs.CL cs.AIcited by 16

On the map: LLM alignment / RLHF / preference optimisation → Safety alignment and jailbreak attacks

also in 1 more
Abstract (arXiv)

Despite substantial advancements in aligning large language models (LLMs) with human values, current safety mechanisms remain susceptible to jailbreak attacks. We hypothesize that this vulnerability stems from distributional discrepancies between alignment-oriented prompts and malicious prompts. To investigate this, we introduce LogiBreak, a novel and universal black-box jailbreak method that leverages logical expression translation to circumvent LLM safety systems. By converting harmful natural language prompts into formal logical expressions, LogiBreak exploits the distributional gap between alignment data and logic-based inputs, preserving the underlying semantic intent and readability while evading safety constraints. We evaluate LogiBreak on a multilingual jailbreak dataset spanning three languages, demonstrating its effectiveness across various evaluation settings and linguistic contexts.

Key facts. Cited by 16 papers in the map (data through 14 Oct 2026). Placed in Safety alignment and jailbreak attacks, one of 10 directions in LLM alignment / RLHF / preference optimisation. Builds on 29 of its 60 references that are in the map. First posted on arXiv 18 May 2025.

Builds on

29 of 60 references.

Closest ancestors · 3

Same direction, newest first

  1. 2024How Alignment and Jailbreak Work: Explain LLM Safety through Intermediate Hidden States9 Jun
  2. 2023GPT-4 Is Too Smart To Be Safe: Stealthy Chat with LLMs via Cipher12 Aug
  3. 2023Universal and Transferable Adversarial Attacks on Aligned Language Models27 Jul

Other references · 26

  1. 2023Jailbroken: How Does LLM Safety Training Fail?cited by 1,651
  2. 2023Jailbreaking Black Box Large Language Models in Twenty Queriescited by 1,420
  3. 2023Large Language Modelscited by 780
  4. 2024JailbreakBench: An Open Robustness Benchmark for Jailbreaking Large Language Modelscited by 536
  5. 2023Jailbreaking ChatGPT via Prompt Engineering: An Empirical Studycited by 498
  6. 2024LLM2Vec: Large Language Models Are Secretly Powerful Text Encoderscited by 404
  7. 2023Low-Resource Languages Jailbreak GPT-4cited by 309
  8. 2023Multilingual Jailbreak Challenges in Large Language Modelscited by 240
  9. 2018Embedding Logical Queries on Knowledge Graphscited by 120
  10. 2024Does Refusal Training in LLMs Generalize to the Past Tense?cited by 79
  11. 2023Harnessing the Power of Large Language Models for Natural Language to First-Order Logic Translationcited by 59
  12. 2015Traversing Knowledge Graphs in Vector Spacecited by 51
14 more
  1. 2024NoteLLM-2: Multimodal Large Representation Models for Recommendationcited by 44
  2. 2024LLM-Powered User Simulator for Recommender Systemcited by 39
  3. 2024Rethinking Large Language Model Architectures for Sequential Recommendationscited by 27
  4. 2024Large Language Model Distilling Medication Recommendation Modelcited by 26
  5. 2024Harnessing Large Language Models for Knowledge Graph Question Answering via Adaptive Multi-Aspect Retrieval-Augmentationcited by 24
  6. 2023MILL: Mutual Verification with Large Language Models for Zero-Shot Query Expansioncited by 23
  7. 2024LLMEmb: Large Language Model Can Be a Good Embedding Generator for Sequential Recommendationcited by 18
  8. 2024Virtual Context: Enhancing Jailbreak Attacks with Special Token Injectioncited by 17
  9. 2020Exploring Neural Models for Parsing Natural Language into First-Order Logiccited by 16
  10. 2024Exploiting the Index Gradients for Optimization-Based Jailbreaking on Large Language Modelscited by 15
  11. 2024GASP: Efficient Black-Box Generation of Adversarial Suffixes for Jailbreaking LLMscited by 12
  12. 2024Improving Complex Reasoning over Knowledge Graph with Logic-Aware Curriculum Tuningcited by 10
  13. 2024Characterizing and Evaluating the Reliability of LLMs against Jailbreak Attackscited by 8
  14. 2025Entailment-Preserving First-order Logic Representations in Natural Language Entailmentcited by 5

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 4; 2026: 12.

16 papers.

Sort:

By area:

  1. 2025Faithful-First Reasoning, Planning, and Acting for Multimodal LLMscited by 7
  2. 2026From Completion to Editing: Unlocking Context-Aware Code Infilling via Search-and-Replace Instruction Tuningcited by 6
  3. 2025Do Methods to Jailbreak and Defend LLMs Generalize Across Languages?cited by 5
  4. 2026Steering to Say No: Configurable Refusal via Activation Steering in Vision Language Modelscited by 3
  5. 2025Scaling Patterns in Adversarial Alignment: Evidence from Multi-LLM Jailbreak Experimentscited by 2
  6. 2026Alignment-Weighted DPO: A principled reasoning approach to improve safety alignmentcited by 2
  7. 2026Type-Checked Compliance: Deterministic Guardrails for Agentic Financial Systems Using Lean 4 Theorem Provingcited by 2
  8. 2025ASGuard: Activation-Scaling Guard to Mitigate Targeted Jailbreaking Attackcited by 1
  9. 2026MOSAIC: Composable Safety Alignment with Modular Control Tokenscited by 1
  10. 2026textbfAGTAO: Robust and Stabilized LLM Unlearning via Adversarial Gating Training with Adaptive Orthogonalitycited by 0
  11. 2026SRTJ: Self-Evolving Rule-Driven Training-Free LLM Jailbreakingcited by 0
  12. 2026Model-Agnostic Lifelong LLM Safety via Externalized Attack-Defense Co-Evolutioncited by 0
4 more
  1. 2026Provably Secure Agent Guardrailcited by 0
  2. 2026Yesterday's Shield, Today's Spear: A Self-Evolving Safety Guardrail in Productioncited by 0
  3. 2026LLM Alignment--Utility Asymmetry under Semantic-Preserving Transformationscited by 0
  4. 2026Controlled Decoding Attacks on Black-Box LLMscited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025Playing the Fool: Jailbreaking LLMs and Multimodal LLMs with Out-of-Distribution Strategy
  2. 2024SafeDecoding: Defending against Jailbreak Attacks via Safety-Aware Decoding
  3. 2025Forewarned is Forearmed: Pre-Synthesizing Jailbreak-like Instructions to Enhance LLM Safety Guardrail to Potential Attacks
  4. 2025Speak Easy: Eliciting Harmful Jailbreaks from LLMs with Simple Interactions
  5. 2024SATA: A Paradigm for LLM Jailbreak via Simple Assistive Task Linkage
  6. 2025SafeInt: Shielding Large Language Models from Jailbreak Attacks via Safety-Aware Representation Intervention
  7. 2025A Domain-Based Taxonomy of Jailbreak Vulnerabilities in Large Language Models
  8. 2025Beyond Jailbreaks: Revealing Stealthier and Broader LLM Security Risks Stemming from Alignment Failures