Introduction

Recent advances in data-driven robot learning have enabled rapid progress in single-arm manipulation through large-scale teleoperation datasets, foundation policies, and vision–language–action models. As these systems begin to exhibit scaling behaviors, a natural question arises: can the same paradigm simply extend to bimanual manipulation, or does dual-arm intelligence require fundamentally new structural inductive biases? 

One line of work suggests that bimanual capability can emerge from scaling data and model capacity, enabled by diverse demonstrations from teleoperation, simulation, and human videos, together with foundation policies. Another perspective argues that bimanual manipulation introduces qualitatively new challenges, including inter-arm coupling, role assignment, temporal coordination, and shared physical constraints, which may require new architectures, coordination mechanisms, or hierarchical representations.

This workshop brings together researchers to examine this scaling–structure question and advance the understanding of bimanual robotic intelligence.

Workshop Schedule

Invited Speakers

Call for Papers

We welcome submissions on recent advances in bimanual manipulation and related problems in robotics and embodied AI. In particular, we encourage contributions spanning approaches that scale data and models for bimanual manipulation, and approaches that develop systems and structures for coordination and control.

Topics of Interest

We welcome contributions including, but not limited to:

Scaling-based approaches for bimanual manipulation

• Large-scale datasets for bimanual manipulation, including teleoperation, simulation, and human demonstrations

• Vision–language–action models and foundation policies for bimanual tasks

• Cross-embodiment learning from human videos for bimanual skill acquisition

Structure-aware approaches for coordination and control

• Coordination mechanisms, role assignment, and interaction modeling for dual-arm systems

• Hierarchical and structured policies for long-horizon bimanual tasks

• Learning-based planning and control for multi-arm manipulation

Bridging scaling and structure

• Hybrid approaches that combine large-scale learning with explicit coordination structures

• Simulation, real-to-sim transfer, and benchmarking for bimanual manipulation

• Evaluation of coordination, physical feasibility, and embodied execution in bimanual systems

Submission Format and Presentation

We welcome both short papers and extended abstracts (up to 4 pages, excluding references) describing ongoing or completed work. Submissions should be made through OpenReview.

Accepted submissions will be presented as poster sessions, with a subset selected for spotlight talks. 

Awards

The workshop will recognize outstanding contributions with the following awards, generously sponsored by PrimeBot:

• 🏆 Best Workshop Paper Award: 1 paper, with a cash prize of USD 1,000.

• 🏆 Outstanding Workshop Paper Award: 3 papers, each with a cash prize of USD 500.

Important Dates

• Submission deadline: August 24, 2026, 11:59 PM AOE

• Acceptance notification: September 6, 2026, 11:59 PM AOE

• Paper camera-ready deadline: September 20, 2026, 11:59 PM AOE

Workshop Organizers