Contact-Anchored Policies: Contact Conditioning Creates Strong Robot Utility Models

Open Pick Close Unseen environments and objects vision action Task success rates by methodFig. 1.We introduce Contact-Anchored Policies (CAP), a method to conditioning multimodal policies with physical contact information.Such policies are able to generalize zero-shot to novel objects and scenes with orders of magnitude less data, compute, and model parameters compared to frontier behavior model, while outperforming them on atomic skills trained with CAP.Abstract-The prevalent paradigm in robot learning attempts to generalize across environments, embodiments, and tasks with language prompts at runtime.A fundamental tension limits this approach: language is often too abstract to guide the concrete physical understanding required for robust manipulation.In this work, we introduce Contact-Anchored Policies (CAP), which replace language conditioning with points of physical contact in space.Simultaneously, we structure CAP as a library of modular utility models rather than a monolithic generalist policy.This factorization allows us to implement a real-tosim iteration cycle: we build EgoGym, a lightweight simulation benchmark, to rapidly identify failure modes and refine our models and datasets prior to real-world deployment.We show that by conditioning on contact and iterating via simulation, CAP generalizes to novel environments and embodiments out of the box on three fundamental manipulation skills while using only 23 hours of demonstration data, and outperforms large, stateof-the-art VLAs in zero-shot evaluations by 56%.All model checkpoints, codebase, hardware, simulation, and datasets will be open-sourced.