Tag: adversarial reinforcement learning
Browse our exclusive articles!
Understanding RL-Jailbreaker Attacks on Large Language Models
Explore how RL-jailbreaker attacks exploit vulnerabilities in LLMs and discover strategies to enhance AI safety against adversarial threats.
f-Divergence Regularized RLHF: Unified Theory & Algorithms
Explore a unified framework for f-divergence regularized RLHF with novel algorithms improving efficiency and performance in reinforcement learning.
Compress KV Cache in RL Post-Training with Shadow Mask
Learn how Shadow Mask Distillation compresses KV cache in RL post-training for memory-efficient alignment and improved learning stability.
Optimizing Adam for Streaming Reinforcement Learning
Explore how revisiting Adam optimizer enhances streaming reinforcement learning with DQN, C51, and Adaptive Q(λ) for stable, efficient AI training.
Gradient Extrapolation-Based Policy Optimization in RL
Discover GXPO, a novel RL method boosting training efficiency and accuracy with faster gradient extrapolation and adaptive optimization.