paper-with-me

홈 › Papers

Reflex-Guard: A Low-Latency Guardrail for LLM Prompt Safety Using Dense Semantic Embeddings

2026-08-18 · Istiaque Ahmed, Afia Anjum Borsha, Ranat Das Prangon, Abu-fuad Ahmad, Thi Hong Tran arxiv

Large Language Models (LLMs) in real-world applications often face the risks of specially crafted prompts designed to bypass the safety controls. Existing guardrail methods, such as LLM-as-a-judge and cloud-based safety APIs are able to detect unsafe content. However, they often add a delay of about 250-900 ms to each request. This delay is too high for real-time applications, when the system usually needs to respond in less than 100 ms. Furthermore, routing user prompts through external moderation endpoints raises significant data privacy concerns. This paper introduces Reflex-Guard, a lightweight guardrail that runs locally. It uses jailbreak-aware preprocessing, compact sentence-transformer embeddings, and seven fast binary classifiers. Together, these components enable high-accuracy prompt safety filtering with much lower latency than existing solutions. Through systematic evaluation on a strategically balanced dataset of 30,568 samples drawn from five complementary sources, we demonstrate that Reflex-Guard achieves 95.9% recall on harmful prompts at 37.6 ms end-to-end latency. It is faster than existing baselines, including Llama Guard 2 at 255 ms and SafeDecoding at 723 ms. It can detect 100% of GCG suffix attacks and Base64-encoded prompts using the default threshold. However, DrAttack structured prompts required lowering the threshold to 0.03 for optimal detection, as they produced a distinct probability distribution. Reflex-Guard achieves Reflex Efficiency Score (RES) scores up to 16.79, significantly outperforming Llama Guard 2 (11.90) and SafeDecoding (9.80). This analysis offers practical deployment advice and shows that different attack types occupy distinct regions in the embedding probability space.

📄 PDF Abstract BibTeX arXiv:2608.17556

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

kNNGuard: Turning LLM Hidden Activations into a Training-Free Configurable Guardrail

2026-07-02 · Mahmoud Abdelfattah, Hamid Nasiri, Peter Garraghan arxiv

Large language models (LLMs) are increasingly deployed in domains requiring guardrails to detect unsafe, off-topic, or adversarial prompts. Existing guardrails predominantly rely on fine-tuning to build classifiers, whic…

Domain Adaptation

DT-Guard: Intent-Driven Reasoning-Active Training for Reasoning-Free LLM Safety Guardrail

2026-07-07 · He Liu, Changtao Miao, Xinjie Yang, Tianle Song 외 arxiv

Large language models deployed in open-world applications require safety guardrails that are both robust to complex risks and efficient enough for low-latency runtime moderation. Existing guardrails face a practical trad…

Auto-Tuning Safety Guardrails for Black-Box Large Language Models

2025-12-14 · Perry Abdulkadir arxiv

Large language models (LLMs) are increasingly deployed behind safety guardrails such as system prompts and content filters, especially in settings where product teams cannot modify model weights. In practice these guardr…

Hyperparameter Optimization

Robust and Efficient Guardrails with Latent Reasoning

2026-05-27 · Siddharth Sai, Xiaofei Wen, Muhao Chen arxiv

Maintaining the safety of large language models (LLMs) is crucial as they are increasingly deployed in real-world applications. Existing safety guardrails typically rely on single-pass classification or, more recently, d…

Evaluating Prompt Injection Defenses for Educational LLM Tutors: Security-Usability-Latency Trade-offs

2026-03-29 · Alexandre Cristovão Maiorano arxiv

Educational LLM tutors face a core AI alignment challenge: they must follow user intent while preserving pedagogical constraints and safety policies. We present an evaluation methodology for prompt-injection defenses in …

Adversarial Robustness