Blogs

How LLMs Scaled from 512 to 2M Context

Feb 12, 2026

Does RLVR Really Improve Reasoning Beyond the Base Model?

Jan 10, 2026

An exploration of why RL with verifiable rewards improves sampling efficiency but may not expand a model’s reasoning capabilities.

San Francisco 2025

Dec 13, 2025

My impression from my first visit to SF.

Neurips 2025

Dec 12, 2025

Learnings from Neurips 2025

Playing to win VS not to lose

Dec 11, 2025

If You Play Not to Lose, Will You Still Win?