SlotGCG uses Vulnerable Slot Score (VSS) to identify and target the most vulnerable prompt positions for adversarial token insertion, delivering 14% higher ASR than standard GCG and 42% higher against defenses.
Safedecoding: Defending against jailbreak attacks via safety-aware decoding
11 Pith papers cite this work. Polarity classification is still indexing.
11
Pith papers citing it
citation-role summary
method 1
citation-polarity summary
roles
method 1polarities
background 1representative citing papers
A multi-turn intention-deception jailbreak achieves high success on GPT-5 and Claude models while exposing para-jailbreaking where models leak harmful information without direct refusal.
HMNS is a new jailbreak method that uses causal head identification and nullspace-constrained injection to achieve higher attack success rates than prior techniques on aligned language models.
SafeSpec integrates a latent safety head into speculative LLM decoding with rollback and reflective multi-sampling, cutting attack success rates 15% on Qwen3-32B while retaining 2.06x speedup on normal workloads.
SafeSteer improves safety in multimodal large language models by up to 33.4% via a decoding probe and modal alignment vector without any fine-tuning.
TrajGuard detects jailbreaks by tracking how hidden-state trajectories move toward high-risk regions during decoding, achieving 95% defense rate with 5.2 ms/token latency across tested attacks.
ReasoningGuard is an inference-time method that uses attention mechanisms to inject safety aha moments and scaling sampling to defend large reasoning models against jailbreak attacks.
Jailbreak attacks suppress Adversarially Compromised Heads in early layers but leave Safety-Aligned Heads active in mid-layers, producing robust harmful features usable for competitive training-free detection.
NeuroArmor uses safe-variant-guided representation consistency checks for selective intervention, reducing jailbreak ASR from 41.56% to 1.57% and benign FPR from 30.26% to 22.05% on Llama-3-8B-Instruct.
A survey that creates taxonomies for jailbreak attacks and defenses on LLMs, subdivides them into sub-classes, and compares evaluation approaches.
citing papers explorer
-
SlotGCG: Exploiting the Positional Vulnerability in LLMs for Jailbreak Attacks
SlotGCG uses Vulnerable Slot Score (VSS) to identify and target the most vulnerable prompt positions for adversarial token insertion, delivering 14% higher ASR than standard GCG and 42% higher against defenses.
-
Jailbreaking Frontier Foundation Models Through Intention Deception
A multi-turn intention-deception jailbreak achieves high success on GPT-5 and Claude models while exposing para-jailbreaking where models leak harmful information without direct refusal.
-
Jailbreaking the Matrix: Nullspace Steering for Controlled Model Subversion
HMNS is a new jailbreak method that uses causal head identification and nullspace-constrained injection to achieve higher attack success rates than prior techniques on aligned language models.
-
SafeSpec: Fast and Safe LLM via Dynamic Reflective Sampling
SafeSpec integrates a latent safety head into speculative LLM decoding with rollback and reflective multi-sampling, cutting attack success rates 15% on Qwen3-32B while retaining 2.06x speedup on normal workloads.
-
SafeSteer: A Decoding-level Defense Mechanism for Multimodal Large Language Models
SafeSteer improves safety in multimodal large language models by up to 33.4% via a decoding probe and modal alignment vector without any fine-tuning.
-
TrajGuard: Streaming Hidden-state Trajectory Detection for Decoding-time Jailbreak Defense
TrajGuard detects jailbreaks by tracking how hidden-state trajectories move toward high-risk regions during decoding, achieving 95% defense rate with 5.2 ms/token latency across tested attacks.
-
ReasoningGuard: Safeguarding Large Reasoning Models with Inference-time Safety Aha Moments
ReasoningGuard is an inference-time method that uses attention mechanisms to inject safety aha moments and scaling sampling to defend large reasoning models against jailbreak attacks.
-
Robust Harmful Features Under Jailbreak Attacks: Mechanistic Evidence from Attention Head Specialization in Large Language Models
Jailbreak attacks suppress Adversarially Compromised Heads in early layers but leave Safety-Aligned Heads active in mid-layers, producing robust harmful features usable for competitive training-free detection.
-
NeuroArmor: Safe-Variant-Guided Representation Consistency for Selective Re-Anchoring in Jailbreak Defense
NeuroArmor uses safe-variant-guided representation consistency checks for selective intervention, reducing jailbreak ASR from 41.56% to 1.57% and benign FPR from 30.26% to 22.05% on Llama-3-8B-Instruct.
-
Jailbreak Attacks and Defenses Against Large Language Models: A Survey
A survey that creates taxonomies for jailbreak attacks and defenses on LLMs, subdivides them into sub-classes, and compares evaluation approaches.
- Minimal, Local, Causal Explanations for Jailbreak Success in Large Language Models