Black et al.: delay-simulated action-prefix conditioning for real-time control

What changed

Simulated inference delay during training and conditioned on action prefixes directly, replacing inference-time inpainting while retaining asynchronous action-chunk execution.

Scientific scope

Simulated inference delay during training and conditioned on action prefixes directly, replacing inference-time inpainting while retaining asynchronous action-chunk execution. Inherits RTC runtime and chunking. Removes additional inpainting overhead, not all model inference or communication latency; costs shift to training/fine-tuning. Simulation and two-task real-world tests do not certify every robot/delay regime.

Evidence

The linked original sources describe the stated contribution and its evidence limits. WRS has not independently rerun the experiments or certified a proof. Inherits RTC runtime and chunking. Removes additional inpainting overhead, not all model inference or communication latency; costs shift to training/fine-tuning. Simulation and two-task real-world tests do not certify every robot/delay regime.

Contributors

• Source-listed authors and technical contributors · SCIENCE: Source-listed authors or explicitly credited technical contributors for the cited versions. Family entries retain source-specific attribution; a combined list is not one original-paper byline or equal individual work. Names alone do not establish resolved identities.

• Training-Time Action Conditioning for Efficient Real-Time Chunking
• Black et al.: delay-simulated action-prefix conditioning for real-time control — institutional account

Provisional WRS assessment, edition 2026-10-16.1. Independent expert calibration is pending. This grade does not certify correctness or assign ownership.

Full scientific record · Browse contribution grades · Request a correction