Vision Language Action using trajectory as action tokens.

Open Research Knowledge Graph (ORKG) 2026 Dataset Open access

Read full text
Authors: Bharatesh Chakravarthi
DOI: 10.48366/r1582158
Cited by: 0

Abstract

Trajectory-based action tokens in VLA research encode actions through motion representations at different granularities. Point Trajectories track the paths of selected keypoints, Visual Trajectories draw explicit motion

Fields of Study & Subjects

Robot Manipulation and Learning Multimodal Machine Learning Applications Reinforcement Learning in Robotics

Publication details

Journal / source
Open Research Knowledge Graph (ORKG)
Publication date
3 February 2026
Language
EN
Document type
Dataset