Controlled Decoding Attacks on Black-Box LLMs
cs.CR, cs.AI
Submitted: 2026-09-29
Updated: 2026-09-29
Code: https://github.com/JessonWong/controlled-decoding
Terminology
Sources
- Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedback
- Jailbreaking Black Box Large Language Models in Twenty Queries
- Accelerating Large Language Model Decoding with Speculative Sampling
- Self-Compression of Chain-of-Thought via Multi-Agent Reinforcement Learning
- The Autonomy Tax: Defense Training Breaks LLM Agents
- FORTIS: Benchmarking Over-Privilege in Agent Skills
- HarmBench: A Standardized Evaluation Framework for Automated Red Teaming and Robust Refusal
- Tree of Attacks: Jailbreaking Black-Box LLMs Automatically
- Logic Jailbreak: Efficiently Unlocking LLM Safety Restrictions Through Formal Logical Expression
- Fine-tuning Aligned Language Models Compromises Safety, Even When Users Do Not Intend To!
- Memory Retrieval for Changing Preferences
- Don't Let It Hallucinate: Premise Verification via Retrieval-Augmented Logical Reasoning
- Great, Now Write an Article About That: The Crescendo Multi-Turn LLM Jailbreak Attack
- JULI: Jailbreak Large Language Models by Self-Introspection
- SORRY-Bench: Systematically Evaluating Large Language Model Safety Refusal
- Learning to Deliberate: Meta-policy Collaboration for Agentic LLMs with Multi-agent Reinforcement Learning
- Auditing Multi-Agent LLM Reasoning Trees Outperforms Majority Vote and LLM-as-Judge
- MIRROR: Learning from the Other View for Multi-Modal Reasoning
- 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