Showing 2026Show all
2 papers · 1 filter
cs.LG2026
Offline RL with Hierarchical Action Chunking
Ahad Jawaid
Offline goal-conditioned reinforcement learning (RL) holds the promise of learning general-purpose policies from static datasets. However, scaling these methods to long-horizon tas…
cs.RO2026
NAC: Neural Action Codec for Vision-Language-Action Models
Ahad Jawaid, Yu Xiang
Vision-language-action (VLA) models rely on discrete action tokenizers to bridge continuous robot control and autoregressive sequence modeling, yet existing tokenizers often trade…