TwinGate deploys a stateful dual-encoder system with asymmetric contrastive learning to detect decompositional jailbreaks in untraceable LLM traffic at high recall and low false-positive rate with negligible latency.
A comprehensive study of jailbreak attack versus defense for large language models
4 Pith papers cite this work, alongside 39 external citations. Polarity classification is still indexing.
4
Pith papers citing it
39
external citations · Crossref
citation-role summary
background 1
method 1
citation-polarity summary
verdicts
UNVERDICTED 4representative citing papers
Steering vectors for refusal primarily modify the OV circuit in attention, ignore most of the QK circuit, and can be sparsified to 1-10% of dimensions while retaining performance.
Targeted prompting and system interventions enable local LLMs such as Llama 3.1 70B to exploit 83% of tested Linux privilege escalation vulnerabilities.
A framework detects LLM anomalies including hallucinations, jailbreaks, and backdoors by forensic inspection of layer-wise hidden state patterns, reporting over 95% accuracy with minimal computational overhead.
citing papers explorer
-
TwinGate: Stateful Defense against Decompositional Jailbreaks in Untraceable Traffic via Asymmetric Contrastive Learning
TwinGate deploys a stateful dual-encoder system with asymmetric contrastive learning to detect decompositional jailbreaks in untraceable LLM traffic at high recall and low false-positive rate with negligible latency.
-
What Drives Representation Steering? A Mechanistic Case Study on Steering Refusal
Steering vectors for refusal primarily modify the OV circuit in attention, ignore most of the QK circuit, and can be sparsified to 1-10% of dimensions while retaining performance.
-
Enhancing Linux Privilege Escalation Attack Capabilities of Local LLM Agents
Targeted prompting and system interventions enable local LLMs such as Llama 3.1 70B to exploit 83% of tested Linux privilege escalation vulnerabilities.
-
Exposing the Ghost in the Transformer: Abnormal Detection for Large Language Models via Hidden State Forensics
A framework detects LLM anomalies including hallucinations, jailbreaks, and backdoors by forensic inspection of layer-wise hidden state patterns, reporting over 95% accuracy with minimal computational overhead.