Paul Zhou's banner

Paul Zhou

@zhiyuan_zhou_ • 1,415 subscribers

RL & robots. phd @berkeley_ai, prev @physical_int

Videos

Anya Rossi

sweetdream.ai

SweetDream.ai•Sponsored•Livecam

Watch Anya Live

Anya is streaming live right now! Join her private show and enjoy exclusive content.

Exclusive private shows

1.2k viewers online

Private Show

Join now for exclusive access

Free preview available • Premium content

Do you ever find finetuning VLA overfits to the target task, to the point where generalist ability is lost and even minor deviations beyond the SFT data break the policy? We found an extremely simple solution: directly merge the base and finetuned policy in weight space 🤯 👇🧵

Do you ever find finetuning VLA overfits to the target task, to the point where generalist ability is lost and even minor deviations beyond the SFT data break the policy? We found an extremely simple solution: directly merge the base and finetuned policy in weight space 🤯 👇🧵

126,621 görüntüleme • 5 ay önce

Can robots self-improve by collecting data autonomously🤖? Introducing SOAR: a system for large-scale autonomous data collection 🚀 and autonomous improvement📈of a multi-task language-conditioned policy in diverse scenes without human interventions .

Can robots self-improve by collecting data autonomously🤖? Introducing SOAR: a system for large-scale autonomous data collection 🚀 and autonomous improvement📈of a multi-task language-conditioned policy in diverse scenes without human interventions .

47,667 görüntüleme • 1 yıl önce

Daha fazla içerik yok.