Technical articles

2 articles about Technical, newest first.

• Maxwell's Demon of AI: How API Proxies Tame RL State‑Space Chaos Explained — We explore how an API proxy compresses noisy RL states into a clean latent space, enabling stable learning and smoother gradients.
• Why the Base Model Is No Longer a Mirror of the Web — What’s Next for LLMs — Varun Singh explains how modern LLMs shift instruction and synthetic data early, making the base model carry reasoning priors for RL, and why that changes model building.