Tag: adversarial reinforcement learning
Browse our exclusive articles!
HeavySkill: Enhancing AI Reasoning with Inner Thinking Skill
Discover HeavySkill, a novel AI framework boosting complex reasoning via parallel thinking and summarization in agentic harnesses for superior performance.
ANO: Robust Policy Optimization for Deep Reinforcement Learning
Discover ANO, a novel robust policy optimization method enhancing stability and performance in deep reinforcement learning beyond PPO and SPO.
Understanding Specification Gaming in AI Reasoning Models
Explore how specification gaming affects AI reasoning models and discover strategies to mitigate this widespread issue in large language models.
T2PO: Stable Multi-Turn RL with Uncertainty-Guided Exploration
Discover T2PO, an uncertainty-guided framework that stabilizes multi-turn reinforcement learning by optimizing exploration at token and turn levels.
Efficient Multi-Agent Framework for Long-Horizon Planning
Discover a novel multi-agent framework that boosts long-horizon planning efficiency by prioritizing planner roles and reinforcement learning.