Karpathy's prediction about RL is coming true now!He called reward functions unreliable and argued that a single reward number is too low-dimensional to teach an agent what "good" means for complex tasks. To solve this, Agents need a knowledge-guided review as a higher-dimensional feedback channel.Every major AI lab trains models with RL today (OpenAI, Anthropic, DeepSeek).And their key bottleneck has always been the reward functions.GRPO by DeepSeek worked well for math and code because the env

In 1963, the mathematician Stanisław Ulam noticed an unusual pattern while doodling in his notebook during a presentation. When integers are arranged in a spiral, prime numbers appear to fall along diagonal lines. At first, this is not entirely surprising, since all prime numbers except 2 are odd, and the diagonals of such spirals alternate between odd and even numbers. What is far more striking is that primes tend to cluster along certain diagonals more than others, regardless of whether the sp

@blancmontagnard @Okabe_Rintar0IF I’m tutoring a girl in linear algebra and I’ve found that insulting and threatening her has gotten her grades up.