Pretrain to Adapt: What Makes a Pretrained Policy Adaptable?

Call for Papers

We invite submissions on all topics related to what makes a pretrained robot policy a good foundation for downstream adaptation. We welcome extended abstracts and short papers on completed work, as well as position papers and work-in-progress submissions that raise open questions.

Topics

Topics of interest include, but are not limited to:

• Pretraining objectives for adaptable policies (behavior cloning, return-conditioned, world-model, value-based, in-context objectives)
• Composing and curating heterogeneous pretraining data (demonstrations, play data, suboptimal/failure trajectories, RL rollouts) for downstream adaptability
• Coupling pretraining recipes with adaptation methods: RL finetuning, in-context learning, and test-time steering
• Steerability and plasticity of pretrained policies
• Adapting pretrained policies to new tasks, embodiments, and environments
• Metrics and benchmarks for measuring readiness to adapt
• World models as pretraining objectives, evaluation tools, or RL environments
• Reward modeling and learned rewards for continual policy improvement
• Position papers, work-in-progress, and retrospectives on any of the above

Submission Guidelines

Deadline

October 4, 2026, 5:00 AM (as shown on OpenReview). Submissions open August 29, 2026.

Page Limit

We welcome short papers of up to 4 pages and full papers of up to 8 pages (both excluding references), describing completed research, ongoing work, or benchmark datasets relevant to the workshop topics.

Formatting

Authors are encouraged to use the CoRL 2026 paper template.

Reviews

Each submission will receive at least two reviews from the organizing committee and external reviewers.

Anonymity

All submissions will be double-anonymized.

Dual Submission

We welcome submissions that are under review or recently published at relevant venues.

Non-Archival

This workshop is non-archival. Accepted papers will be posted on the workshop website.

Presentation

Accepted papers will be presented in an interactive poster session, with the top 3–5 papers selected for oral spotlight presentations. Authors are expected to present in person.

Submissions will be handled through OpenReview.

Submit via OpenReview

Important Dates

Times as shown on OpenReview.