🌻 Renovated my blog. Now I can have a cup of coffee. ☕ — Introduction Previous version of my blog focused on experimental and visual look. Since the first blog post in August 2017, thinking ...

Humanioids conference review — Towards force intelligence, from position intelligence. Of course, there were some robots dancing. Great keynote speech by Shuran Song. Referen...

What is the the most important unit of the transformer — Let’s look at the intuitive visualization) (LLM Visualization — Bbycroft.net, n.d.) . An interactive visualization of an LLM. The ...

Reaction to Seohong's Post on X — The Search for Scalable Q Learning I have long searched for ways to scale up Reinforcement Learning. One interesting project of such is the starcraft.ai pro...

Reading Jensen Huang's Biography. — Who is Jensen Huang? Currently reading the book “Thinking Machines” (Witt et al., 2025) , a biography of the NVIDIA’s CEO, Jensen Huang. I noticed this b...

💻 Imaginary yet inspiring conversation between Richard Feynman and Bill Gates, generated by the recent version of ChatGPT. — Quantum Question: While reading (Nielsen & Chuang, 2011) , I...

One of the must-see papers of 2022. Finally, but as expected, it's here: decision-making formulated not as reinforcement learning (RL), but as conditional generative modeling. Personally, I am doub...

A single graph neural network processor capable of learning to execute a wide range of algorithms, such as sorting, searching, dynamic programming, path-finding, and geometry. — [\renewcomman...

🧠 A generalist agent proposed by DeepMind, a multi-modal, multi-task, multi-embodiment generalist policy inspired from large-scale language modeling — Let’s take a look at a generalist agent ...