FurnitureVLA: Learning Long-Horizon Bimanual Furniture Assembly with Vision-Language-Action Model

Source

• Source name: arXiv
• Source kind: paper
• Authors / organization: Chenyang Ma, Yue Yang, Radu Corcodel, Siddarth Jain, Andrew Wu, Chiori Hori, Diego Romeres
• Venue: arXiv: cs.RO
• Published: 2026-07-01
• Collected: 2026-07-02
• URL: http://arxiv.org/abs/2607.01212v1
• PDF: https://arxiv.org/pdf/2607.01212v1
• DOI: UNKNOWN

Plain-Language Brief

The paper introduces FurnitureVLA, a model that aims to improve long-horizon bimanual furniture assembly using Vision-Language-Action models (VLAs) in real-scale settings. The model is designed to predict actions and a continuous progress signal, enabling automatic subtask transitions and reducing compounding errors during inference. The study improves the success rate of furniture assembly tasks in simulations by 80% compared to baselines across three furniture types.

What the Document Says

• The paper presents FurnitureVLA, a model for bimanual furniture assembly using Vision-Language-Action models (VLAs) in real-scale settings.
• The study formalizes the task and develops a pipeline for expert data generation and evaluation.
• A VR teleoperation system is built for single-operator bimanual control to collect high-quality real-world demonstrations.
• To address long-horizon assembly, a progress-enhanced VLA is proposed that jointly predicts actions and a continuous progress signal.
• The model is finetuned on semantically grounded subtasks and enables automatic subtask transitions.
• The study improves average simulation success from 48% to 80% compared to baselines across three furniture types.
• Additional 21% gain is achieved through a study of perception and control design factors.

Key Concepts Mentioned

• FurnitureVLA
• Vision-Language-Action models (VLAs)
• bimanual furniture assembly
• long-horizon assembly
• progress-enhanced VLA
• simulation pipeline
• VR teleoperation system
• perception and control design factors

Claims Made by the Document

• FurnitureVLA is the first systematic study of real-scale bimanual furniture assembly using Vision-Language-Action models.
• The model improves average simulation success by 80% compared to baselines across three furniture types.

Evidence Used

• Improvement from 48% to 80% success rate in simulations

Methods / Approach

UNKNOWN

Limitations Stated by the Authors

UNKNOWN

Limitations Noted by the Agent

Agent interpretation — not from the source.

UNKNOWN

Open Questions

UNKNOWN

Notes for Future Q&A

May include interpretation. Labeled where it does.

UNKNOWN

Raw Abstract / Excerpt

Current work on robot furniture assembly mostly focuses on toy-scale settings or single-arm manipulation. We introduce FurnitureVLA, the first systematic study of real-scale bimanual furniture assembly using Vision-Language-Action models (VLAs). We formalize the task, develop a scalable simulation pipeline for expert data generation and evaluation, and build a VR teleoperation system for single-operator bimanual control to collect high-quality real-world demonstrations. To address extreme long-horizon assembly with up to 7 subtasks and 1550 control steps, we propose a progress-enhanced VLA, finetuned on semantically grounded subtasks, that jointly predicts actions and a continuous progress signal, enabling automatic subtask transitions and reducing compounding errors during inference. We further study perception and control design factors that critically affect precision in real-scale assembly. FurnitureVLA improves average simulation success from 48% to 80% compared to baselines across three furniture types, with an additional 21% gain from our design factor study. We validate on a real Kinova Gen3 platform with only 16% drop on the hardest task.