Cheng Chi(@chichengcc):The spatial segregation of learning vs classical robotics expertise often sorts people into camps by where they went to school. But the growing intersection of the two is where the real-world performance bottleneck lives. The people who globally optimize the entire system will win in the end. Kudos to @younghyo_park and @BronarsToni for pushing past the false dichotomy that dominates academic framing.

2026.04.08 22:18

The spatial segregation of learning vs classical robotics expertise often sorts people into camps by where they went to school. But the growing intersection of the two is where the real-world performance bottleneck lives. The people who globally optimize the entire system will win in the end. Kudos to @younghyo_park and @BronarsToni for pushing past the false dichotomy that dominates academic framing.

Younghyo Park@younghyo_park

2026.04.08 12:59

What's different between these two BC policies? It's the same architecture, training budget, and data collection setup — the only difference is the controller gains! Controller gains are an understudied design parameter in robot learning. In our new work (w/ @BronarsToni*, @pulkitology), we show how they act as an inductive bias across BC, RL, and Sim2Real transfer, with real consequences on performance. Here's what we found 🧵 * Equal Contribution 📄arxiv: 🔗website: