
Model family
MAWAM
Participating organizations
Participating organizationsMIT
Participating organizationsNVIDIAParticipating organizationsThe University of Hong KongVersion scope and roles for these associations await verification.
Inputs include a language instruction, camera histories, robot states, embodiment metadata, roles, and view ownership. A shared simulation camera can supply read-only context. Each prediction covers 24 control steps and eight future frames per private view. Robots execute eight steps in sync and replan. The authors report 67.8% across three real tasks (30 trials each) and 90.3% across eight RoboFactory tasks (100 episodes each). Table 9 reports 2.76 billion trainable parameters, excluding frozen Cosmos-Reason1 and the video VAE.
Overview
- Model type
- Action policy
- Architecture
- WAM · Transformer
- Action generation
- Flow matching
- Listed versions
- MAWAM
- Related robots
- 1
Papers & reports (1)
MAWAM: A Multi-Agent World-Action Model for Coordinated Manipulation
Paper available
arXiv: 2611.25105DOI: 10.48550/arxiv.2611.25105
Authors & affiliations (13)
Author order: Original order
- First published
- 2026-11-20
- Latest paper revision
- v1
- Venue
- Not provided
Covered model versions
Official resources
- NVIDIA MAWAM project ↗
Publisher not provided · Source
Publication date unknown · Accessed: 2026-11-20
Submit a source or correction