Reposted by @faridlazuarda.bsky.social
🚨 Paper Alert: “RL Finetunes Small Subnetworks in Large Language Models”
From DeepSeek V3 Base to DeepSeek R1 Zero, a whopping 86% of parameters were NOT updated during RL training 😮😮
And this isn’t a one-off. The pattern holds across RL algorithms and models.
🧵A Deep Dive
