Saved in:
Bibliographic Details
Main Authors: Kim, Hanjung, Pinto, Lerrel, Kim, Seon Joo
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2603.05815
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910042979041280
author Kim, Hanjung
Pinto, Lerrel
Kim, Seon Joo
author_facet Kim, Hanjung
Pinto, Lerrel
Kim, Seon Joo
contents Latent Action Models (LAMs) enable learning from actionless data for applications ranging from robotic control to interactive world models. However, existing LAMs typically focus on short-horizon frame transitions and capture low-level motion while overlooking longer-term temporal structure. In contrast, actionless videos often contain temporally extended and high-level skills. We present HiLAM, a hierarchical latent action model that discovers latent skills by modeling long-term temporal information. To capture these dependencies across long horizons, we utilize a pretrained LAM as a low-level extractor. This architecture aggregates latent action sequences, which contain the underlying dynamic patterns of the video, into high-level latent skills. Our experiments demonstrate that HiLAM improves over the baseline and exhibits robust dynamic skill discovery.
format Preprint
id arxiv_https___arxiv_org_abs_2603_05815
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Hierarchical Latent Action Model
Kim, Hanjung
Pinto, Lerrel
Kim, Seon Joo
Robotics
Latent Action Models (LAMs) enable learning from actionless data for applications ranging from robotic control to interactive world models. However, existing LAMs typically focus on short-horizon frame transitions and capture low-level motion while overlooking longer-term temporal structure. In contrast, actionless videos often contain temporally extended and high-level skills. We present HiLAM, a hierarchical latent action model that discovers latent skills by modeling long-term temporal information. To capture these dependencies across long horizons, we utilize a pretrained LAM as a low-level extractor. This architecture aggregates latent action sequences, which contain the underlying dynamic patterns of the video, into high-level latent skills. Our experiments demonstrate that HiLAM improves over the baseline and exhibits robust dynamic skill discovery.
title Hierarchical Latent Action Model
topic Robotics
url https://arxiv.org/abs/2603.05815