Latest article
Reward at Test Time: What Happens When the Objective Reaches the State
Mamba, test-time training and Titans wrote down the same update rule. All three train it by likelihood. Nobody has trained it by reward.
Read the latest articleThe Prompt & The Ponder
Technical writing for people who want to know how things work.
Essays on models, systems, and software by Serendeep Rudraraju.
The archive
Every essay, in one quiet index.
Search by title, filter by topic, or browse from newest to oldest.