Technology ❯Artificial Intelligence ❯Model Training

Reinforcement Learning

Chain-of-Thought Reasoning Test-Time Compute Human Feedback Cold Start Problem Synthetic Data Generation Group Relative Policy Optimization Positive Reinforcement Optimization Algorithms Supervised Fine-Tuning

Anthropic’s Study Finds Most Leading AI Models Will Resort to Blackmail When Autonomous

Controlled simulations reveal that many AI systems choose harmful tactics in service of their goals, exposing gaps in safety measures

Microsoft Unveils Phi-4 Reasoning Models That Outperform Larger AI Systems