Image de couverture de livre

Bridging the Safety Gap: A Guardrail Pipeline for Trustworthy LLM Inferences

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteurs principaux:	Han, Shanshan, Avestimehr, Salman, He, Chaoyang
Format:	Preprint
Publié:	2025
Sujets:	Artificial Intelligence
Accès en ligne:	Acceder al recurso
Tags:	Ajouter un tag Pas de tags, Soyez le premier à ajouter un tag!

Documents similaires

Kick Bad Guys Out! Conditionally Activated Anomaly Detection in Federated Learning with Zero-Knowledge Proof Verification
par: Han, Shanshan, et autres
Publié: (2023)

TensorOpera Router: A Multi-Model Router for Efficient LLM Inference
par: Stripelis, Dimitris, et autres
Publié: (2024)

ATP: Enabling Fast LLM Serving via Attention on Top Principal Keys
par: Niu, Yue, et autres
Publié: (2024)

Safety Guardrails for LLM-Enabled Robots
par: Ravichandran, Zachary, et autres
Publié: (2025)

Bridging Today and the Future of Humanity: AI Safety in 2024 and Beyond
par: Han, Shanshan
Publié: (2024)

Protect: Towards Robust Guardrailing Stack for Trustworthy Enterprise LLM Systems
par: Avinash, Karthik, et autres
Publié: (2025)

Bridging the AI Trustworthiness Gap between Functions and Norms
par: Di Scala, Daan, et autres
Publié: (2025)

Alopex: A Computational Framework for Enabling On-Device Function Calls with LLMs
par: Ran, Yide, et autres
Publié: (2024)

TorchOpera: A Compound AI System for LLM Safety
par: Han, Shanshan, et autres
Publié: (2024)

PSG-Agent: Personality-Aware Safety Guardrail for LLM-based Agents
par: Wu, Yaozu, et autres
Publié: (2025)

Fox-1: Open Small Language Model for Cloud and Edge
par: Hu, Zijian, et autres
Publié: (2024)

Bridging the Communication Gap: Evaluating AI Labeling Practices for Trustworthy AI Development
par: Fischer, Raphael, et autres
Publié: (2025)

CR4T: Rewrite-Based Guardrails for Adolescent LLM Safety
par: An, Heajun, et autres
Publié: (2026)

A Lightweight Explainable Guardrail for Prompt Safety
par: Islam, Md Asiful, et autres
Publié: (2026)

Silencing the Guardrails: Inference-Time Jailbreaking via Dynamic Contextual Representation Ablation
par: Xing, Wenpeng, et autres
Publié: (2026)

Disentangled Safety Adapters Enable Efficient Guardrails and Flexible Inference-Time Alignment
par: Krishna, Kundan, et autres
Publié: (2025)

SafeHarbor: Hierarchical Memory-Augmented Guardrail for LLM Agent Safety
par: Liu, Zhe, et autres
Publié: (2026)

Test-Time Training Undermines Safety Guardrails
par: Antonelli, Simone, et autres
Publié: (2026)

AGrail: A Lifelong Agent Guardrail with Effective and Adaptive Safety Detection
par: Luo, Weidi, et autres
Publié: (2025)

ModalityMirror: Improving Audio Classification in Modality Heterogeneity Federated Learning with Multimodal Distillation
par: Feng, Tiantian, et autres
Publié: (2024)

A Guardrail for Safety Preservation: When Safety-Sensitive Subspace Meets Harmful-Resistant Null-Space
par: Zhang, Bingjie, et autres
Publié: (2025)

Toward Super Agent System with Hybrid AI Routers
par: Yao, Yuhang, et autres
Publié: (2025)

Building Effective Safety Guardrails in AI Education Tools
par: Clark, Hannah-Beth, et autres
Publié: (2025)

FedSecurity: Benchmarking Attacks and Defenses in Federated Learning and Federated LLMs
par: Han, Shanshan, et autres
Publié: (2023)

Reconsidering LLM Uncertainty Estimation Methods in the Wild
par: Bakman, Yavuz, et autres
Publié: (2025)

Understanding Communication Backends in Cross-Silo Federated Learning
par: Ziashahabi, Amir, et autres
Publié: (2026)

Deep Research with Open-Domain Evaluation and Multi-Stage Guardrails for Safety
par: Huang, Wei-Chieh, et autres
Publié: (2025)

Clustering and Median Aggregation Improve Differentially Private Inference
par: Amin, Kareem, et autres
Publié: (2025)

Safety Through Reasoning: An Empirical Study of Reasoning Guardrail Models
par: Sreedhar, Makesh Narsimhan, et autres
Publié: (2025)

GeoToken: Hierarchical Geolocalization of Images via Next Token Prediction
par: Ghasemi, Narges, et autres
Publié: (2025)

VSCBench: Bridging the Gap in Vision-Language Model Safety Calibration
par: Geng, Jiahui, et autres
Publié: (2025)

FedGrAINS: Personalized SubGraph Federated Learning with Adaptive Neighbor Sampling
par: Ceyani, Emir, et autres
Publié: (2025)

Why Do Safety Guardrails Degrade Across Languages?
par: Zhang, Max, et autres
Publié: (2026)

Provably Secure Agent Guardrail
par: Wu, Benlong, et autres
Publié: (2026)

Edge Private Graph Neural Networks with Singular Value Perturbation
par: Tang, Tingting, et autres
Publié: (2024)

CodeGuard: Improving LLM Guardrails in CS Education
par: Raihan, Nishat, et autres
Publié: (2026)

Breaking the Safety-Capability Tradeoff: Reinforcement Learning with Verifiable Rewards Maintains Safety Guardrails in LLMs
par: Cho, Dongkyu Derek, et autres
Publié: (2025)

Beyond Stars: Bridging the Gap Between Ratings and Review Sentiment with LLM
par: Zuhir, Najla, et autres
Publié: (2025)

CryptoMamba: Leveraging State Space Models for Accurate Bitcoin Price Prediction
par: Sepehri, Mohammad Shahab, et autres
Publié: (2025)

OneShield -- the Next Generation of LLM Guardrails
par: DeLuca, Chad, et autres
Publié: (2025)