Safety Critical Task

Safety-critical tasks, demanding reliable performance with minimal risk of catastrophic failure, are a central focus in artificial intelligence research. Current efforts concentrate on developing robust reinforcement learning algorithms, including model-based and model-free approaches, often incorporating safety constraints through techniques like projection, recovery mechanisms, or human intervention. These methods aim to improve the safety and reliability of AI systems in diverse applications, from robotics and autonomous driving to large language models, addressing challenges such as uncertainty quantification and out-of-distribution detection. The ultimate goal is to create trustworthy AI systems capable of handling high-stakes situations while adhering to strict safety protocols.

Papers

October 23, 2024

MobileSafetyBench: Evaluating Safety of Autonomous Agents in Mobile Device Control
Juyong Lee, Dongyoon Hahm, June Suk Choi, W. Bradley Knox, Kimin Lee
Autonomous Agent Mobile Device Safety Guarantee Mobile User Safety Evaluation Remote Control Safety Critical Task

October 22, 2024

Insights on Disagreement Patterns in Multimodal Safety Perception across Diverse Rater Groups
Charvi Rastogi, Tian Huey Teh, Pushkar Mishra, Roma Patel, Zoe Ashwood, Aida Mostafazadeh Davani, Mark Diaz, Michela Paganini, Alicia Parrish, Ding Wang, Vinodkumar Prabhakaran, Lora Aroyo, Verena Rieser
Generative AI DCU Insight AQ Human Rating Safety Climate Perception Multimodal Control High Disagreement Multi Rater Safety Critical Task

March 4, 2024

Offline Goal-Conditioned Reinforcement Learning for Safety-Critical Tasks with Recovery Policy
Chenyang Cao, Zichen Yan, Renhao Lu, Junbo Tan, Xueqian Wang
Reinforcement Learning Recovery Guarantee FETCH Robot Offline Safe Reinforcement Learning Offline Goal Conditioned Reinforcement Learning Safety Critical Task

February 8, 2024

In-Context Learning Can Re-learn Forbidden Tasks
Sophie Xhonneux, David Dobre, Jian Tang, Gauthier Gidel, Dhanya Sridhar
Large Language Model Context Learning LLM Generated LLM Safety Knowledge Unlearning Safety Critical Task

August 6, 2023

Building Safe and Reliable AI systems for Safety Critical Tasks with Vision-Language Processing
Shuang Ao
Artificial Intelligence Uncertainty Quantification Model Uncertainty Accurate Uncertainty System Safety Safety Critical Task

July 28, 2023

TrafficSafetyGPT: Tuning a Pre-trained Large Language Model to a Domain-Specific Expert in Transportation Safety
Ou Zheng, Mohamed Abdel-Aty, Dongdong Wang, Chenzhu Wang, Shengxuan Ding
Domain Expertise Safety Critical Task

December 12, 2022

Evaluating Model-free Reinforcement Learning toward Safety-critical Tasks
Linrui Zhang, Qin Zhang, Li Shen, Bo Yuan, Xueqian Wang, Dacheng Tao
Reinforcement Learning Safe Reinforcement Learning Safety Constraint Probabilistic Safety Safety Layer Safety Critical Task

November 10, 2021

Look Before You Leap: Safe Model-Based Reinforcement Learning with Human Intervention
Yunkun Xu, Zhenyu Liu, Guifang Duan, Jiangcheng Zhu, Xiaolong Bai, Jianrong Tan
Deep Reinforcement Learning Model Predictive Control Leap Forward Real World System Model Based Safe Reinforcement Learning Safety Critical Task

Safety Critical Task

Papers

MobileSafetyBench: Evaluating Safety of Autonomous Agents in Mobile Device Control

Insights on Disagreement Patterns in Multimodal Safety Perception across Diverse Rater Groups

Offline Goal-Conditioned Reinforcement Learning for Safety-Critical Tasks with Recovery Policy

In-Context Learning Can Re-learn Forbidden Tasks

Building Safe and Reliable AI systems for Safety Critical Tasks with Vision-Language Processing

TrafficSafetyGPT: Tuning a Pre-trained Large Language Model to a Domain-Specific Expert in Transportation Safety

Evaluating Model-free Reinforcement Learning toward Safety-critical Tasks

Look Before You Leap: Safe Model-Based Reinforcement Learning with Human Intervention