Research papers
Submissions addressing system-level challenges in scalable robot learning.
Machine Learning/CORE Unranked
ScaleInfra@IROS2026 2026
IROS 2026 Workshop on Building Scalable Infrastructure for Robot Learning: From Data Scaling to Real-World Deployment
The ScaleInfra@IROS2026 workshop focuses on building scalable infrastructure for robot learning, addressing system-level challenges from data acquisition to real-world deployment. It aims to bridge algorithmic innovation with engineering practice across robotics, machine learning, and computer vision communities, emphasizing reproducible evaluation, continuous learning, and reusable infrastructure.
Paper fit
Contribution paths
A strong submission should clearly identify its contribution and evaluate it appropriately.
System and infrastructure reports
Detailed accounts of engineering systems and infrastructure developed for robot learning.
Benchmark or evaluation papers
Papers proposing or analyzing benchmarking protocols and evaluation methodologies.
Work in progress
Early-stage research or prototypes seeking community feedback.
Negative or failure-case studies
Studies documenting failures or challenges that provide practical insights for the community.
Research areas in scope
01
Key Topics & Discussion Pillars
Data Scaling: scalable strategies for acquiring, generating, and orchestrating multimodal data across real-world interactions and simulationsTraining and Post-Training: supervised fine-tuning, imitation learning, reinforcement learning, preference learning, and scalable training workflows for embodied agentsEvaluation: standardized benchmarking protocols for generalization, long-horizon reasoning, safety, robustness, and data efficiencyDeployment Infrastructure: reliable software-hardware integration pipelines and continuous feedback loops for real-world robot execution and fleet learning
02
Scope of Accepted Papers
Real-world robot interaction data collection and large-scale teleoperation pipelinesSimulation data generation, synthetic multimodal data, and sim-to-real transferSupervised fine-tuning, imitation learning, and post-training workflows for VLA modelsOffline and online reinforcement learning, and preference learning for embodied agentsBenchmarking VLA capabilities, long-horizon reasoning, and task generalizationSafety, robustness assessment, and reliable evaluation methodologies for real robotsInfrastructure for continuous deployment, hardware-software integration, and fleet learningDistributed experimentation and runtime diagnostics for real-world robot execution systems
Policies worth checking twice
- Submissions will be reviewed using a double-blind process.
- Submissions must use the IROS / IEEE two-column conference template.
- Papers must not exceed 8 pages, including references.
- Accepted submissions will not be included in official proceedings.
- Previously published work may be submitted for discussion with the workshop community.
Official sources
Compiled from the official call for papers. The organizers’ pages remain authoritative.
Last verified September 9, 2026