← Core AI
Artificial Intelligence › Core AI ›

Safety

317 directly classified papers

Papers per year

Papers

Constrained Update Projection Approach to Safe Policy Optimization NIPS 2022
Two Coupled Rejection Metrics Can Tell Adversarial Examples Apart CVPR 2022
Stability Verification in Stochastic Control Systems via Neural Network Supermartingales AAAI 2022
SaFeRDialogues: Taking Feedback Gracefully after Conversational Safety Failures ACL 2022
Tight Neural Network Verification via Semidefinite Relaxations and Linear Reformulations AAAI 2022
Distillation of RL Policies with Formal Guarantees via Variational Abstraction of Markov Decision Processes AAAI 2022
Safe Online Convex Optimization with Unknown Linear Safety Constraints AAAI 2022
DeformRS: Certifying Input Deformations with Randomized Smoothing AAAI 2022
Conservative and Adaptive Penalty for Model-Based Safe Reinforcement Learning AAAI 2022
Preemptive Image Robustification for Protecting Users against Man-in-the-Middle Adversarial Attacks AAAI 2022
Exploring Safer Behaviors for Deep Reinforcement Learning AAAI 2022
Leashing the Inner Demons: Self-Detoxification for Language Models AAAI 2022
‘Beach’ to ‘Bitch’: Inadvertent Unsafe Transcription of Kids’ Content on YouTube AAAI 2022
Handling and Presenting Harmful Text in NLP Research EMNLP 2022
ProsocialDialog: A Prosocial Backbone for Conversational Agents EMNLP 2022
Red Teaming Language Models with Language Models EMNLP 2022
SafeText: A Benchmark for Exploring Physical Safety in Language Models EMNLP 2022
Planning to Avoid Side Effects AAAI 2022
Best Arm Identification with Safety Constraints AISTATS 2022
On the Safety of Conversational Models: Taxonomy, Dataset, and Benchmark ACL 2022
Counterfactual harm NIPS 2022
Probabilistic Robustness Quantification of Neural Networks AAAI 2021
Safe Policy Optimization with Local Generalized Linear Function Approximations NIPS 2021
Tightening Robustness Verification of Convolutional Neural Networks with Fine-Grained Linear Approximation AAAI 2021
Transfer Learning for Efficient Iterative Safety Validation AAAI 2021