action representation 1benchmark 1diffusion models 1embodied ai 1human following 1language-guided navigation 1optical flow 1robotic manipulation 1vision-language models 1world models 1
From the 2 of 2 linked papers with an AI index.
2 papers
cs.AI2026
UESF-Bench: Benchmarking and Probing for Unified Embodied Seeking and Following
Kun Yu, Jianhua Yang, Yixiang Chen +7
The paper introduces UESF-Bench, a large-scale benchmark for unified embodied seeking and following of humans, and presents SeekFollow-VLA, a vision‑language‑action framework that…
cs.RO2026
FlowWAM: Optical Flow as a Unified Action Representation for World Action Models
Yixiang Chen, Peiyan Li, Yuan Xu +13
The paper introduces FlowWAM, a dual‑stream diffusion model that uses optical flow as a unified video‑native representation of actions, enabling both action prediction and world mo…