DayNews.ai

RL algebra shift boosts AI reasoning

Swapping addition for maximum in RL math lets AI agents compose reasoning steps from separate attempts.