ideasthesia

Sign in Subscribe

Ryan Collison

I write about cool science and share it here.

Synthesis of inference scaling insights about thinking

Synthesis: What Inference Scaling Teaches About the Nature of Thinking

Series: Test-Time Compute Scaling | Part: 9 of 9 We started with a simple observation: language models that think longer produce better answers. We end with something deeper: a formal theory of what thinking is. Test-time compute scaling isn’t just an engineering trick for better AI. It’s

Test-time compute as active inference free energy minimization

Test-Time Compute Meets Active Inference: Reasoning as Free Energy Minimization

Series: Test-Time Compute Scaling | Part: 8 of 9 Here’s the deep connection: test-time compute scaling and active inference are describing the same process from different angles. One says: Intelligence scales with how thoroughly you search reasoning space. The other says: Intelligence is minimizing prediction error through inference.

Pay-per-intelligence business model visualization

The Economics of Inference: Pay-Per-Intelligence Business Models

Series: Test-Time Compute Scaling | Part: 7 of 9 When intelligence becomes tunable—when you can dial up thinking time to improve output quality—the economics of AI fundamentally change. The old model: pay per token, pay per API call. Intelligence is fixed. Everyone gets the same capability regardless of

Model self-verification and refinement loop

Self-Refinement and Verification: Models That Check Their Work

Series: Test-Time Compute Scaling | Part: 6 of 9 Humans don’t just produce answers—we check them. We read our own writing, debug our own code, verify our own math. We catch errors, revise, refine, and iterate until we’re confident the answer is right. This self-checking loop

Monte Carlo tree search applied to language model reasoning

Tree Search in Language Models: Monte Carlo Meets GPT

Series: Test-Time Compute Scaling | Part: 5 of 9 Monte Carlo Tree Search (MCTS) revolutionized game-playing AI. It’s how AlphaGo beat Lee Sedol at Go, how chess engines achieve superhuman play, how autonomous systems plan complex sequences of actions. The algorithm is simple: explore a tree of possible

Trade-off visualization between training and inference compute

The Compute Trade-Off: When to Train vs When to Think

Series: Test-Time Compute Scaling | Part: 4 of 9 Here’s the new strategic question facing AI developers: you have a fixed compute budget. Do you spend it on pretraining a bigger model, or on letting a smaller model think longer at inference time? This wasn’t a question before.

Extended chain of thought reasoning process

Chain of Thought on Steroids: The Mechanics of Extended Reasoning

Series: Test-Time Compute Scaling | Part: 3 of 9 When researchers discovered chain-of-thought prompting in 2022, it seemed like a clever trick. Ask a model to “think step by step,” and it performs better on reasoning tasks. Simple. Elegant. Moderately effective. But what o1 does isn’t chain-