3 papers
cs.LG2026
FASTER: Value-Guided Sampling for Fast RL
Perry Dong, Alexander Swerdlow, Dorsa Sadigh +1
Some of the most performant reinforcement learning algorithms today can be prohibitively expensive as they use test-time scaling methods such as sampling multiple action candidates…
cs.LG2026
TQL: Scaling Q-Functions with Transformers by Preventing Attention Collapse
Perry Dong, Kuo-Han Hung, Alexander Swerdlow +2
Despite scale driving substantial recent advancements in machine learning, reinforcement learning (RL) methods still primarily use small value functions. Naively scaling value func…
cs.LG2025
: a VLA That Learns From Experience
Physical Intelligence, Ali Amin, Raichelle Aniceto +53
We study how vision-language-action (VLA) models can improve through real-world deployments via reinforcement learning (RL). We present a general-purpose method, RL with Experience…