Full papers
Submissions up to 8 pages, with references and appendix not counting toward the page limit; must be double-blind and use any academic paper template.
Machine Learning/CORE Unranked
CSIR 2026
CoRL 2026 Workshop on Continually Self-Improving Robots
The Continually Self-Improving Robots (CSIR) 2026 workshop, co-located with CoRL 2026 in Austin, Texas, focuses on advancing robotic systems that can autonomously improve their capabilities beyond initial demonstrations through interaction, data collection, and learning. It brings together researchers in robot learning, reinforcement learning, and continual learning to address foundational challenges in formalizing self-improvement and building practical pipelines for long-term robotic autonomy.
Paper fit
Contribution paths
A strong submission should clearly identify its contribution and evaluate it appropriately.
Research areas in scope
01
Topics of Interest
Robot self-improvementContinual robot learningRL for roboticsAutonomous data collectionFailure discoverySafe explorationWorld modelsIn-context robot learningReal-to-sim-to-real transferAutomatic reset systemsReliable evaluation environments
Policies worth checking twice
- Submissions must be double-blind: remove author names, affiliations, and other identifying information.
- Papers are limited to 8 pages (references and appendix excluded).
- Accepted papers are non-archival and will not appear in archival proceedings.
- Any academic paper template may be used for submissions.
Official sources
Compiled from the official call for papers. The organizers’ pages remain authoritative.
Last verified September 9, 2026